Food Chemistry 143 (2014) 139–146

Contents lists available at ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Effect of a Thermoascus aurantiacus thermostable enzyme cocktailon wheat bread qualitiy

0308-8146/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.foodchem.2013.07.103

⇑ Corresponding author. Tel.: +55 17 3221 2267; fax: +55 17 3221 2299.E-mail addresses: denisefronteira@gmail.com (D.S. Oliveira), javier@ibilce.

unesp.br (J. Telis-Romero), dasilva@ibilce.unesp.br (R. Da-Silva), celia@ibilce.unesp.br (C.M.L. Franco).

D.S. Oliveira a, J. Telis-Romero a, R. Da-Silva b, C.M.L. Franco a,⇑a Laboratory of Cereals, Roots, and Tubers, UNESP – São Paulo State University, Rua Cristovão Colombo, 2265 São José do Rio Preto, São Paulo CEP 15054-000, Brazilb Laboratory of Biochemistry and Applied Microbiology, UNESP – São Paulo State University, Rua Cristovão Colombo, 2265 São José do Rio Preto, São Paulo CEP 15054-000, Brazil

a r t i c l e i n f o a b s t r a c t

Article history:Received 20 February 2013Received in revised form 3 July 2013Accepted 20 July 2013Available online 27 July 2013

Keywords:Thermoascus aurantiacus CBMAI 756XylanaseArabinoxylanBread stalingAmylopectin retrogradationCrumb firmness

Thermophilic fungus Thermoascus aurantiacus (CBMAI 756) on solid-state fermentation using corncob asa nutrient source produces an enzyme pool with the potential to be used in bread making. In this paper,the use of this enzyme cocktail as a wheat bread improver was reported. Both products released by flourarabinoxylan degradation and bread quality were investigated. The main product released throughenzyme activity after prolonged incubation was xylose indicating the presence of xylanase; however, asmall amount of xylobiose and arabinose also confirmed the presence of xylosidase and a-L-arabinofura-nosidase, respectively. Enzyme mixture ‘‘in vitro’’ mainly attacked water-unextractable arabinoxylan con-tributing to beneficial effect in bread making. The use of an optimal enzyme concentration (35 Uxylanase/100 g of flour) increased specific volume (22%), reduced crumb firmness (25%), and reducedamylopectin retrogradation (17%) during bread storage. In conclusion, the enzyme cocktail producedby T. aurantiacus CBMAI 756 can improve wheat bread quality.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Bread staling is a complex phenomenon that happens duringstorage, and it is largely caused by water migration and transfor-mations that occur in the starch. Alterations related to this phe-nomenon include an increase in moisture in the crust (loss ofcrispiness), an increase in crystallinity in the starch granule, an in-crease in crumb firmness, a loss of organoleptic properties in theloaves, and the crumb’s loss of water-holding capacity (Gray &BeMiller, 2003; Ribotta & Le Bail, 2007). For years, it was believedthat amylopectin retrogradation was the most important phenom-enon responsible for the increase in crumb firmness during breadstorage. Nevertheless, bread staling does not occur only becauseof amylopectin retrogradation (Gray & Bemiller, 2003; Martin,Zeleznak, & Hoseney, 1991). Gluten-starch interactions and mois-ture transfer also seems to be involved in bread staling (Gray &Bemiller, 2003).

In most countries, the bread is made from wheat flour. The mainconstituents of flour are starch and proteins, although minor com-pounds, such as lipids and non-starch polysaccharides (NSP)(including arabinoxylan (AX) and b-glucans), also influence theprocess and quality of the final product. Among NSPs, wheat AX

has been reported to be the most important. Cereal AX is classifiedinto water-extractable arabinoxylan (WE-AX) and water-unex-tractable arabinoxylan (WU-AX). AX concentration in wheat flourvaries between 1.5% and 2.5%, among which 0.4–0.6% is WE-AX(Courtin & Delcour, 2002).

In cereals, these polymers are comprised of a principal chainmade up of b-D-xylopyranosyl units, linked through 1,4 glycosidiclinkages that can be substituted in various degrees by a-L-arabino-furanosyl residues at positions C-2 and/or C-3 of the xylose(Kulkarni, Shendye, & Rao, 1999). Other substituents, such asglucuronic acid, D-galactose, and phenol groups (ferulic acid, andp-coumaric acid), may also be present (Subramaniyan & Prema,2002). Due to the great complexity of AX structure, a cocktail of en-zymes containing endo-1,4-b-xylanase (EC 3.2.1.8), 1,4-b-xylosi-dase (EC 3.2.1.37), a-L-arabinofuranosidase (EC 3.2.1.55),acetylxylan esterase (EC 3.1.1.72), feruloyl esterase (EC 3.1.1.73),and p-coumaric esterase (EC 3.1.1.-), which synergistically act tohydrolyze this heteropolysaccharide, is required for its completedegradation (Subramaniyan & Prema, 2002; Waters, Murray, Ryan,Arendt, & Tuohy, 2010).

Bread-making industries have used diverse agents to reducebread staling, including enzymes, which have received specialattention. Amylases, proteases and hemicellulases have beenreported as having a direct influence on starch retrogradationand final product quality (Goesaert, Slade, Levine, & Delcour,2009; Verjans, Dornez, Delcour, & Courtin, 2010; Waters, Ryan,Murray, Arendt, & Tuohy, 2011).

140 D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146

Because of the hydrolytic action on non-starch polysaccharides,xylanase has received special attention, and this enzyme is com-monly found in formulations for bread improvement. It has bene-ficial effects on bread quality – it improves rheological propertiesof dough, the specific volume of bread, the crumb texture, andthe bread staling rate (Damen et al., 2012; Laurikainen, Härkönen,Autio, & Poutanen, 1998; Martinez-Anaya & Jiménez, 1997; Shah,Shah, & Madamwar, 2006; Waters et al., 2011).

According to Courtin and Delcour (2001, 2002), xylanases thatpreferentially attack the WU-AX positively impact both the proper-ties of the dough and the volume of the bread, because they affectmoisture redistribution in the system and break glycosidic linkageswith consequent water liberation and reduction in WU-AX size.Thus, the significant challenge today is to obtain xylanases thatwill act on WU-AX, modifying its structure and functionality.

Most studies have shown mesophilic xylanase activity, but littlecan be found regarding the use of thermophilic xylanases. Theadvantage of using heat-stable enzymes comes from a longer activ-ity during the baking period. Studies using recombinant thermo-philic endoxylanase in breads showed an increase in specificvolume, and a reduction in crumb firmness and less amylopectinretrogadation during the storage of these products (Jiang, Bail, &Wu, 2008; Jiang, Li, Yang, Li, & Tan, 2005). According to Jianget al. (2005), thermophilic enzyme can maintain activity duringgreat part of baking period, enhancing the breakdown of AX atincreasing temperature resulting in more oven spring during bak-ing. However, for hyperthermophilic enzymes, such as those activeafter 100 �C, a caution should be taken since an excessive degrada-tion of AX can negatively alter dough properties and bread quality.

Recently, thermophilic fungi have attracted growing attentionin baking, and enzyme cocktails from Talaromyces emersonii havebeen successfully applied to bread making (Waters et al., 2010,2011). The thermophilic fungus Thermoascus aurantiacus, a not tox-in-producing microorganism (Kongbuntad, Saenphet, Khanong-nuch, & Lumyong, 2006; Oliveira, Meherb-Dini, Franco, Gomes, &Da-Silva, 2010), has been used to produce xylanase on solid-statefermentation (SSF) (Da-Silva et al., 2005). In a recent study inwhich corncob was used as a substrate, T. aurantiacus expresseda cocktail of hemicellulolitic enzymes that contained high xylanaseand low protease and amylase activities. This enzyme profile fitswell into the baking industry, because it resulted in low degrada-tion of starch and gluten, but showed high solubilization of xylan(Oliveira et al., 2010).

The effect of xylanases on the bread staling rate is still not com-pletely known; thus, the goal of this study was to investigate theinfluence of the enzyme cocktail with xylanolytic activity from T.aurantiacus fungus CBMAI 756 on bread quality and the stalingprocess, as well as to identify the specific products releasedthrough its activity on wheat flour arabinoxylans. To the authors’knowledge, this is the first report dealing with T. aurantiacus en-zymes as a bread improver.

2. Materials and methods

2.1. Material

Wheat flour with 72% extraction rate was kindly donated by theSete Irmãos Mill (Uberlândia, Brazil). The wheat flour contained0.48% ash, 1.43% fat, 11.78% protein, and 3.50% total dietary fiber,which were determined according to the American Association ofCereal Chemists methods (AACC, 2000). Farinograph and extensog-raph parameters of the wheat flour, determined according to theAACC methods 54–21 and 54–10 (AACC, 2000), respectively, were:water absorption capacity (63.9%), stability time (10 min), mixture

tolerance index (30 BU), extension resistance (520 BU), extensibil-ity (135 mm), and proportional number (3.8).

The other ingredients that were used to prepare the loaves wereobtained from a local market (São José do Rio Preto, Brazil). Chro-matographic degree standards, arabinose (A), and xylose (X1) wereacquired from Fluka (Madrid, Spain), and glucose (G1) was ob-tained from Sigma Aldrich (St. Louis, USA). Xylooligosaccharidestandards with DP 2–6 (X2–X6) were acquired from Megazyme(Bray, Ireland). The a-amylase enzyme from Aspergillus oryzaewas kindly supplied by Granotec (Curitiba, Brazil), and the amylo-glucosidase was isolated from genetically modified Saccharomycescerevisiae in the Laboratory of Biochemistry and Applied Microbiol-ogy of São Paulo State University (São José do Rio Preto, Brazil). Allother chemicals used were of analytical grade.

2.2. Methods

2.2.1. Microorganism, culture conditions and enzyme extractMicroorganism and enzyme production was carried out as pre-

viously described by Oliveira et al. (2010). Briefly, the fungus T.aurantiacus CBMAI 756 was cultivated at 50 �C and kept in testtubes with slanting Sabouraud dextrose agar (Oxoid Ltd., Basing-stoke, Hampshire, England) at room temperature. Mycelial suspen-sion in mineral solution was used to inoculate 250-mL Erlenmeyerflasks containing 10 g of corncob as the nutrient source for SSF(67% moisture) at 50 �C for 6 days. The crude enzyme solutionwas produced by adding 40 mL of distilled water to the fermentedmaterial, and it was concentrated with 75% ethanol at 4 �C over-night. The precipitates were separated using centrifuge at13,700g for 10 min and were then dissolved in a small amount(4 mL) of distilled water.

2.2.2. Enzyme activitiesAll the enzyme reactions and their controls were run in tripli-

cate. The activities of xylanolitic enzymes (endo-1,4-b-xylanase –EC 3.2.1.8 and 1,4-b-xylosidase – EC 3.2.1.37), endoglucanase orCMCase (EC 3.2.1.4), avicelase (EC 3.2.1.91) and a-amylase (EC3.2.1.1) were determined by the incubation of 0.1 mL of enzymaticsolution (appropriately diluted) with 0.9 mL of acetate buffer0.10 mol/L (pH 5.0) The solution contained 0.5% Birchwood xylan(Sigma) in the case of the xylanase and xylosidase combination;it contained 2% carboxymethylcellulose (CMC-Sigma C5768) inthe case of endoglucanase; it contained 1% cellulose (avicel-Sigma)in the case of avicelase, and it contained 0.5% soluble starch (Mal-linckrodt) in the case of a-amylase. After incubating the reactionmixture at 60 �C for 10 min, 1 mL of dinitrosalicylic acid (DNS)was added to interrupt the reaction. The reducing sugars releasedfrom the reaction were determined according to Miller (1959).Controls were prepared by adding the enzyme after the DNS. OneInternational Unit (IU) of enzymatic activity was defined as onemicromole of reducing substances expressed as xylose (xylanaseand xylosidase) or glucose (CMCase, avicelase, a-amylase) releasedper minute under the aforementioned test conditions using stan-dard curves of either xylose or glucose. The activity of amylogluco-sidase was determined by the peroxidase/glucose-oxidaseenzymatic method (Bergmeyer & Bernt, 1974).

The b-glucosidase activity (EC 3.2.1.21) was determined byincubating 50 lL of enzymatic extract in 250 lL of acetate buffer0.1 mol/L (pH 5.0) and 250 lL of p-nitrophenyl-b-D-glucopyrano-side 0.004 mol/L (PNPG, Sigma) (Palma-Fernandez, Gomes, &Da-Silva, 2002). The reaction mixture was kept at 60 �C for10 min. The enzymatic reaction was stopped by the addition of2 mL of sodium carbonate 2 mol/L, and the released nitrophenolwas quantified using a spectrophotometer (Jasco, V-639 Bio, Brazil)at 410 nm. One unit of enzymatic activity was defined as thequantity of the enzyme needed to release 1 lmol of nitrophenol

D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146 141

per minute of reaction, using a standard curve of p-nitrophenol(Sigma). The 1,4-b-xylosidase (EC 3.2.1.37) and a-L-arabinofura-nosidase (EC 3.2.1.55) were determined in the same manner asb-glucosidase, except that the substrates were p-nitrophenyl-b-D-xylopyranoside (PNPX, Sigma) and p-nitrophenyl-a-L-arabinopyr-anoside (PNPA, Sigma), respectively. Protease activity was deter-mined by TCA (Trichloroacetic acid) method according to Oliveiraet al. (2010).

2.2.3. Isolation of arabinoxylan from wheat flourWU-AX was isolated following Courtin, Roelants, and Delcour

(1999), with modifications. Wheat flour (250 g) was added to160 mL of water and mixed for 15 min in a planetary mixer (Lieme,BP-12SL, Caxias do Sul, Brazil). The dough was hand-washed in2500 mL of distilled water, and the resulting starch milk was cen-trifuged at 5000g for 10 min at 15 �C. From the precipitate, a darklayer called the squeegee fraction (SQF), which contains the WU-AX, starch, and proteins, was manually separated from the layerof prime starch. The SQF was dissolved in 250 mL of water and cen-trifuged at 10,000g for 10 min at 15 �C. This extraction process wasrepeated many times in order to obtain the largest quantities ofSQF (400 g). The recovered SQF (400 g) was suspended in an ace-tate buffer 0.1 mol/L (pH 4.5, 1:10 w/v) and heated at 90 �C for10 min to gelatinize the starch. A mixture of a-amylase(10,000 U) and amyloglucosidase (80 U) was added to the sample,which was incubated at 50 �C for 48 h to degrade the residualstarch. After incubation, the sample was filtered through a 170-lm membrane and centrifuged at 10,000g for 15 min at 15 �C.The precipitate was washed six times in water to remove the prod-ucts from the hydrolysis, as well as the soluble proteins. The puri-fied sample was boiled for 15 min to inactivate the enzymes,centrifuged (as described above), and the precipitate was washedtwice with absolute ethanol and dried in an air circulation ovenat 32 �C.

WE-AX was isolated as described by Frederix, Van, Courtin, andDelcour (2004), with modifications. Wheat flour (200 g) washeated at 130 �C for 90 min in an air circulation oven in order toinactivate endogenous enzymes. The flour was dispersed in water(1:10 w/v on fresh weight basis) at room temperature. It was stir-red continuously for 90 min, and was then centrifuged at 3000g for15 min. The supernatant was recovered and heated at 90 �C for30 min in order to precipitate soluble proteins and to gelatinizethe starch. It was again centrifuged at 10,000g for 15 min. ThepH of the supernatant was adjusted to 4.5 using HCl 0.01 mol/L.A mixture of a-amylase (7.000 U) and amyloglucosidase (120 U)was added to the sample, which was incubated at 50 �C for 12 hto degrade any residual starch. After this period, the reaction mix-ture was heated in order to inactivate enzymes, and was centri-fuged at 10,000g for 15 min. A part of the supernatant (100 mL)was mixed with 0.2% of bentonite (10 mL), and this mixture wasstirred continuously at room temperature for 30 min. Next, themixture was centrifuged at 10,000g for 20 min and the supernatantwas filtered using a Whatmam (n� 1) filter paper, and was thenlyophilized.

2.2.4. Hydrolysis of the WU-AX and WE-AXWU-AX and WE-AX were incubated with T. aurantiacus

CBMAI 756 enzyme cocktail. The amount of enzyme cocktailused was calculated with a base in its xylanase activity (UX).Xylanase (35 UX) was added to both AX (3 mg/mL) in acetatebuffer 0.1 mol/L (pH 5.0) and incubated at 32 �C and 70 �Cduring 0, 0.5, 2, 4, 6, 8, 24 and 48 h while being stirredcontinuously. The temperatures of 32 and 70 �C were used be-cause they are the dough fermentation and enzyme stabilitytemperatures, respectively. After the incubation periods, sampleswere centrifuged at 9000g for 10 min. The recovered supernatant

was then boiled by 10 min to inactivate the enzyme and wasagain centrifuged (as described above). Products obtained fromthe hydrolysis were kept frozen for later identification and quan-tification by high-performance anion-exchange chromatography.

2.2.5. High-performance anion-exchange chromatography with pulsedamperometric detection (HPAEC-PAD)

The products from the hydrolysis of WU-AX and WE-AX ob-tained as described in item 2.2.4 were analyzed using an HPAEC-PAD system (ICS 3000, Dionex Corporation, Sunnyvale, USA)equipped with an automatic sampler (AS-40, Dionex Corporation,Sunnyvale, USA) according to the procedure used by Rantanenet al. (2007), with modifications. All samples were filtered (0.22-lm membrane) and injected into the HPAEC-PAD system (20-lLsample loop). The flow rate was 1 mL min�1 at 35 �C. The standardquadruple potential waveform was employed with the followingperiods and pulse potentials: E1 = 0.10 V (t1 = 0.40 s); E2 = �2.00 V(t2 = 0.02 s); E3 = 0.60 V (t3 = 0.01 s); and E4 = �0.10 V (t4 = 0.06 s).All eluents were prepared with ultrapure water (18 mO cm) withN2 sparging. The xylooligosaccharides were separated using a Dio-nex CarboPac™ PA-100 guard-column (4 � 50 mm) and a DionexCarboPac™ PA-100 column (4 � 250 mm). Eluent A was NaOH0.2 mol/L, eluent B was ultrapure water, and eluent C was sodiumacetate 0.5 mol/L in NaOH 0.15 mol/L. The first isocratic phase was5 min in 50% A and 50% B, followed by a linear gradient from 50% Aand 50% B to 44% A and 16% C over the course of 20 min. Xylobiose(X2) (Megazyme) and Xylooligosaccharies (X3–X6) (Megazyme)were used as references. The monosaccharides were separatedusing a Dionex CarboPac™ PA-1 guard-column (4 � 50 mm) anda CarboPac™ PA-1 (4 � 250 mm, Dionex, Sunnyvale, USA) columnwith the same eluents and under the same conditions mentionedabove. An isocratic elution was applied (5% A, 95% B for 20 min).Glucose (G1), arabinose (A), and xylose (X1) were used as chro-matographic standards. The data was analyzed using the Chrome-leon software, version 6.8 (Dionex Corporation, USA). The analysiswas performed in duplicate.

2.2.6. Bread-makingBread dough was formulated with the following ingredients in

fresh weight relative to the amount of wheat flour: dehydratedyeast (Saccharomyces cerevisiae) (2%), salt (1.5%), sugar (4%), hydro-genated vegetable fat (3%) and water (60%). The amount of the T.aurantiacus CBMAI 756 enzyme cocktail was calculated with a basein its xylanase activity (UX), its main compound. Bread made with-out exogenous enzymes served as the control. For loaves producedwith the addition of the enzyme at concentrations of 20, 35 and50 UX xylanase/100 g of flour, the enzyme cocktail was dissolvedin the water used during initial preparations to guarantee its accessto all the dough during the mixing process. The concentrationsused in the work were chosen after preliminary tests based inthe specific volume of breads obtained. The flour, sugar, vegetablefat and most of the water (80%) were mixed for 1 min in a plane-tary mixer (Lieme, BP-12SL, Caxias do Sul, Brazil). The rest of thewater (20%) and the yeast were then added to the dough andmixed for 1 min. Finally, the salt was added to the dough andmixed for 15 min for gluten development. The dough was left toset for 10 min, was divided into 100 g portions and was then leftto set for another 15 min before being rolled and mechanicallyshaped (Universo, MQ, São Paulo, Brazil) and placed in baking pans(15 � 8 cm). The dough was then proofed at 32 �C for 1 h andbaked at 180 �C in an industrial oven (Pasiani, Itajobi, Brazil) for10 min. After baking, the loaves were cooled for 2 h at room tem-perature, packaged in polyethylene bags, and then stored at 4 �Cfor 0, 3, 5, 7, and 10 days for staling studies.

142 D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146

2.2.7. Specific volume of the breadThe bread volume was measured 2 h after baking using rape-

seed displacement according to the AACC method 10–05.01 (AACC,2000). The bread’s specific volume was calculated as the ratio be-tween volume and weight (mL/g).

2.2.8. Water content in crumb and crustCrumb and crust moisture contents of the loaves stored for dif-

ferent days were analyzed using an infrared moisture analyzer(AND, model MX-50, Japan). About 1 g each of crumb and crustwere used for the analysis. Three replicates of all samples wereanalyzed.

2.2.9. Firmness of the bread crumbThe firmness of the crumb of the loaves stored for different days

was determined using a Texture Analyzer (TA-XT2i, Stable Micro-systems, Surrey, UK) according to the AACC method 74–09 (AACC,2000). Samples of breadcrumb (25 mm thickness and 20 mmdiameter) were obtained using a metal molder. The maximumforce (Newton) needed to compress a sample of breadcrumb to40% of its original height using a probe of 25 mm in diameter at5 s intervals between compressions was defined as the firmnessvalue of the crumb. Ten replicates of all samples were analyzed.

2.2.10. Differential scanning calorimetryCalorimetric measurements of the crumb of the loaves stored

for different days were determined using a Differential ScanningCalorimeter (DSC, Pyris 1 Perkin Elmer, Fremont, USA). Breadcrumb samples (3 mg, dry weight basis) were weighed in smallaluminum DSC pans, mixed with 9 lL of deionized water andsealed in the DSC’s accessory apparatus. The pans were kept atroom temperature for 2 h and scanned at a rate of 5 �C/min from25 to 125 �C. An empty aluminum pan was used as a reference.The enthalpy of amylopectin retrogradation was calculated usingthe Pyris 1 software (Perkin Elmer, Fremont, USA). The analysiswas performed in triplicate.

2.2.11. Statistical analysisThe software Statistics for Windows (v. 5.0, Statsoft, Tulsa, USA)

was used to analyze mean values using analysis of variance (ANO-VA). Differences were evaluated using the t-test with Tukey’sadjustment. The significance level was set at a p value <0.05.

3. Results and discussion

3.1. Enzyme activities

The enzyme cocktail used in this work was produced by solidstate fermentation with corncob as a substrate and with 6 daysof incubation. The study of the extracellular amylolytic, proteolytic,cellulolytic, and hemicellulolytic enzymes showed that T.aurantiacus was more xylanolytic than amylolytic, proteolytic,and cellulolytic. The enzymatic profile of the alcoholic precipitated(75%) presented the following composition of activities, in U/mL:Xylanase (endo-1,4-b-xylanase) 600; 1,4-b-xylosidase 5; a-L-arab-inofuranosidase 10; CMCase 340; b-glucosidase 38; protease 90.Avicelase, a-amylase and amyloglucosidase were <1 U. These re-sults are in agreement with previous works (Da-Silva et al., 2005;Oliveira et al., 2010).

3.2. Hydrolysis of WU-AX and WE-AX from wheat flour using enzymecocktail

Products obtained from the hydrolysis of AX extracted fromwheat flour and with the enzyme cocktail were identified and

quantified using HPAEC-PAD. The peaks of the identified productsafter enzymatic hydrolysis of the WU-AX are shown in chromato-grams in Figs. 1 and 2. Disaccharide xylobiose (X2), followed by oli-gosaccharides xylotriose (X3) and xylotetrose (X4), were allidentified (Fig. 1). Xylooligosaccharides with a higher degree ofpolymerization (DP) were not detected. Swennen, Courtin,Bruggen, Vandecasteele, and Delcour (2005) detected xylooligosac-charides with a DP up to 6 when working with Aspergillus aculeatusthermophilic xylanase and using WU-AX isolated from flour assubstrate. Fig. 1 also shows xylose (X1) coeluted with glucose(G1). Its separation and quantification was only possible whenthe samples were analyzed in a PA-1 chromatographic column spe-cific for monosaccharide. Then, arabinose (A), glucose (G1), andxylose (X1) were identified (Fig. 2) evidencing the presence of thedifferent AX-hydrolyzing enzymes and debranching enzyme inthe enzymatic cocktail used. The same reaction products were de-tected when WE-AX was used as substrate; however, the detectedpeaks were lower than those observed for WU-AX (data notshown). The results of this study are in agreement with thosereported by Rantanen et al. (2007) who, when working with xylan-ases fromAspergillus aculeatus thermophilic, detected A, X1, X2 andX3 as products of the AX enzymatic hydrolysis.

Enzymatic activity on WU-AX gradually increased the arabinose(A) peak area during the incubation period (Figs. 1 and 2), mainlyat a temperature of 70 �C, confirming the presence of a-L-arabino-furanosidade (debranching xylanase) in the enzymatic cocktailstudied. When the WE-AX substrate was used, the arabinose peakarea was lower than when WU-AX was used (data not shown).Debranching enzymes are important in the AX hydrolysis process,because the branches restrict the access of xylanases to the 1,4linkages of the principal chain (Martinez-Anaya & Jiménez,1997). Wheat flour AX is highly branched made up of an averageof 35–40% arabinose substituent. a-L-arabinofuranosidases differas to the substrate specificity. Some act only on a-1,2 or a-1,3 gly-cosidic linkages of the mono-substituted xylose residues, whileothers only release arabinose joined by a-1,3 linkages in disubsti-tuted xylose residues (Courtin & Delcour, 2001, 2002). In thisstudy, the specificity of the a-L-arabinofuranosidases found in theenzymatic cocktail produced by T. aurantiacus CBMAI 756 wasnot studied.

Fig. 2 also reflects the pronounced increase in the glucosepeak area over the course of incubation. The enzymatic cocktailproduced by T. aurantiacus CBMAI 756 did not have any amylo-lytic enzymes (a-amylase and glucoamylase), as also observed ina previous study (Oliveira et al., 2010). This increase in theglucose content may be attributed to the degradation of somecellulose present in the AX from the wheat flour. Also, glucosemay be a product of hydrolysis of b(1–3)-D-glucan, howeverb(1–3)-glucanase was not determined in the cocktail. The pres-ence of other hemicellulosic enzymes in the partially purifiedenzymatic cocktail was verified. The concentration of carboxym-ethylcellulase (CMCase) was 340 U/mL in the precipitated enzy-matic cocktail, but after normalization for 35 UX, it was 19 U,and the concentration of b-glycosidase was 38 U/mL (2.1 U afternormalization for 35 UX). The production profile of X1 and X2

from xylanolytic enzyme on WU-AX and WE-AX over the course48 h of incubation, the production of these carbohydrates hadstill not reached a stationary level. The concentration of theseproducts was higher when the xylanase was incubated at70 �C. When the substrate was WE-AX, the concentration ofthe same products (X1 and X2) was considerably lower, andthe production of X1 reached a stationary level after 24 h ofincubation. In this experiment, the temperature did not influenceenzyme activity.

The analysis performed in this study indicated that the enzy-matic cocktail is mainly comprised of endo-1,4-b-xylanases (EC

Fig. 1. HPAEC-PAD profiles of the degradation of WU-AX incubated with enzymecocktail produced by T. aurantiacus CBMAI 756 at 70 �C at different times. A:arabinose; X1: xylose; X2: xylobiose; X3: xylotriose e X4: xylotetraose. Analyticalcolumn PA-100.

Fig. 2. HPAEC-PAD profiles of the degradation of WU-AX incubated with enzymecocktail produced by T. aurantiacus CBMAI 756 at 70 �C at different times. A:arabinose; G1: glucose; X1: xylose. Analytical column PA-1.

D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146 143

3.2.1.8); however lower activities of 1,4-b-xylosidases (EC3.2.1.37), and a-L-arabinofuranosidases (EC 3.2.1.55) were alsopresent. The T. aurantiacus CBMAI 756 enzyme cocktail alsoshowed a significant effect on the WU-AX from wheat flour(Fig. 3) suggesting a potential effect in bread making. WU-AX neg-atively affects dough and bread characteristics while WE-AX andhigh molecular weight solubilized AX have a positive effect(Courtin & Delcour, 2001, 2002; Courtin et al., 1999).

3.3. Specific volume of the bread

The specific volume of loaves made without and with the addi-tion of Thermoascus’s enzyme cocktail varied from 3.46 to4.21 mL/g. There was a significant increase (p 6 0.05) in the specificvolume of the bread when the enzyme cocktail was added at a con-centration of 35 U xylanase (UX)/100 g of flour (4.21 mL/g). How-ever, at concentrations of 20 and 50 UX/100 g of flour, with

Fig. 3. Concentration of xylose (A) and xylobiose (B) obtained from the degradation of WU-AX and WE-AX incubated at 32 and 70 �C in function of time, detected by HPAEC-PAD. WU-AX 32 �C (-s-), WU-AX 70 �C (-d-), WE-AX 32 �C (-h-), WE-AX 70 �C (- j -).

144 D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146

specific volumes of 3.75 and 3.68 mL/g, respectively, there were nosignificant differences in specific volume when compared to thecontrol loaves (3.46 mL/g, without the enzyme cocktail). A concen-tration of 20 UX/100 g of flour was likely not enough to degradethe AX to an optimum degree of degradation. A higher concentra-tion of the enzyme (50 UX/100 g of flour) resulted in a dough thatwas incapable of retaining the CO2 produced by the yeast duringfermentation. Damen et al. (2012) also observed that high concen-trations of xylanase did not increase the specific volume in breads.According to McCleary (1986), excessive hydrolysis of the AX canresult in a weakening of the gluten, thus producing moist andsticky dough. Loaves that had thermostable xylanase from T. marí-tima MSB8 added also showed higher specific volumes than thoseof the control bread (Jiang et al., 2005, 2008). Similar results werealso reported by Laurikainen et al. (1998), who noted an increase of20% in the loaf’s specific volume when xylanase was added to thedough. This value is similar to that found in this study (22%) when35 UX of xylanase per 100 g of flour was used. However, Keskin,Sumnu, and Sahin (2004) did not find any change to the specificvolume in the loaves produced with commercial xylanase. The po-sitive or negative influence of xylanase on the specific volume ofthe loaves is likely related to the quantity and macromolecularcharacteristics of AX in the wheat flour, along with the specificityof the enzymes for the different substrates. The WU-AX is capableof negatively affecting the formation of gluten because of the phys-ical interference of these polysaccharides during the formation ofintra and/or intermolecular interactions of gliadin and gluteninproteins (Courtin et al., 1999; Martinez-Anaya & Jiménez, 1997).Therefore, xylanase that degrades WU-AX, such as that shown inthis study, could help in the formation of a more flexible and stabledough that would have a greater ability to expand during baking.An increase in resistance to extension and a light decrease in

Table 1Moisture content of bread crumb and crust with and without Thermoascus aurantiacus CB

Xylanase (UX/100 g of flour) Crumb Moisture (g/100 g)

Day 0 Day 3

0 40.12 ± 0.48aA 36.62 ± 0.55aB

20 38.61 ± 0.83aA 34.81 ± 0.49bB

35 40.28 ± 0.77aA 35.59 ± 0.38abB

50 40.23 ± 0.50aA 35.32 ± 0.71abB

Crust Moisture (g/100g)0 24.25 ± 0.59aC 27.65 ± 0.42aB

20 23.69 ± 0.31aC 27.17 ± 0.65aB

35 24.17 ± 0.34aC 27.97 ± 0.36aAB

50 23.72 ± 0.51aC 27.42 ± 0,45aB

Values are presented as mean value ± standard deviation (n = 3). Values followed by theline are significantly different from the Tukey test (p 6 0.05), for each analysis.

extensibility of dough were observed in samples that had enzymecocktail added (data not shown).

Using different carbon sources (green tea leaves, wheat bran,wheat flour, sorghum, and glucose), Waters et al. (2010) inducedthe thermophilic fungus Talaromyces emersonii to produce five en-zyme cocktails and used them in bread making. They observed asignificant increase in bread volumes (7–23%) when the enzymecocktail had xylanase. In bread that was infused with the enzymecocktail that had only amylase enzymes, there was a 6% reductionin bread volume. These results indicate that AX degradation, to cer-tain degree may be related to increased CO2 retention capacity, aspartially solubilized WU-AX of HMW enhances the viscosity ofcontinuous phase of the dough. Verjans et al. (2010) attributedthe increase of the specific volume in breads that had xylanasefrom Aureobasidium pullulans added to limited solubilization ofWU-AX with production of HMW molecules during mixture, fer-mentation and baking process.

3.4. Moisture content

Table 1 shows the moisture contents of the crumb and crust ofthe loaves with or without Thermoascus’s enzyme cocktail afterstorage at 4 �C for up to 10 days. The moisture of fresh bread is di-rectly related to its softness. The ideal moisture level of the crumbin fresh breads is between 35% and 40% (Shah et al., 2006). In thisstudy, the crumb moisture content of fresh loaves with or withoutenzyme cocktail was approximately 40%, and on the tenth day ofstorage, no further significant difference (p > 0.05) in the crumbmoisture content was found between the control loaves and thosewhich had enzyme added – all had a satisfactory moisture contentof approximately 35%. However, the crust moisture content of all ofthe breads increased until the third day, remained unchanged until

MAI 756 enzyme cocktail after storage at 4 �C.

Day 5 Day 7 Day 10

35.47 ± 0.20aC 35.62 ± 0.13aBC 36.31 ± 0.52aBC

35.20 ± 0.19abB 34.35 ± 0.49bB 34.79 ± 0.87aB

35.71 ± 0.39aB 34.48 ± 0.38bB 35.11 ± 0.62aB

34.50 ± 0.30bBC 33.48 ± 0.45bC 35.97 ± 0.85aB

28.39 ± 0.39abB 28.05 ± 0.20abB 29.58 ± 0.48aA

27.37 ± 0.60abB 27.16 ± 0.32bB 29.38 ± 0.64aA

27.14 ± 0.60bB 28.40 ± 0.30aA 28.66 ± 0.13ªA

28.70 ± 0.62aAB 28.80 ± 0.67aAB 29.75 ± 0.74aA

different lowercase in the same column, and the different capital letter in the same

Table 2Retrogradation enthalpy (DH) of bread amylopectins with and without Thermoascus aurantiacus CBMAI 756 enzyme cocktail stored at 4 �C during 10 days.

Xylanase (UX/100 g of flour) DH (J/g)

Day 0 Day 3 Day 5 Day 7 Day 10

0 nd 1.89 ± 0.05aB 2.30 ± 0.13abA 2.29 ± 0.06aA 2.10 ± 0.08abAB

20 nd 1.92 ± 0.04aC 2.14 ± 0.01bcA 2.10 ± 0.04abAB 1.97 ± 0.09bBC

35 nd 1.95 ± 0.03aA 2.06 ± 0.04cA 2.04 ± 0.10bA 2.05 ± 0.02abA

50 nd 1.95 ± 0.12aB 2.47 ± 0.03aA 2.33 ± 0.11aA 2.26 ± 0.13aA

Values are presented as mean value ? standard deviation (n = 3). Values followed by the different lowercase in the same column, and by the differentcapital letter in the same line are significantly different from the Tukey test (p 6 0.05); nd: not detected.

D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146 145

the seventh day, and on the tenth day, it was approximately 29%.This reduction in crumb moisture and increase in crust moistureduring storage indicated that there was a migration of water fromcrumb to crust. Similar results were found by Jiang et al. (2008),who used xylanase from Thermotoga maritima MSB8 in breadsstored for 5 days at 4 �C. Though present in the flour in small quan-tities, AX is an important functional ingredient. It absorbs nearlyten times its weight in water, which represents nearly 30% of thewater absorption capacity of wheat flour, thus considerably affect-ing the quality of the dough and bread (Courtin & Delcour, 2002).In this study, the water absorption capacity of wheat flour in-creased with the enzyme cocktail concentration used (data notshown). The preferential action of the enzyme cocktail on WU-AX can have improved the redistribution of water in the dough,however considerable differences were not found between themoisture contents of the crumb and crust of the control loavesand the loaves with the enzyme added.

3.5. Crumb firmness and amylopectin retrogradation

Fig. 4 shows crumb firmness of breads with and without the en-zyme cocktail and stored at 4 �C for up to 10 days. The crumb firm-ness of all loaves studied increased progressively until the tenthday. The rate of this increase was lower after the fifth day of stor-age. There was practically no texture profile difference betweenthe control loaves and those with 20 UX/100 g of flour, except atthe end of the storage period, when the loaves with 20 UX had aslight reduction in crumb firmness (4%). This concentration ofxylanase was not enough to cause a significant increase in thebread’s specific volume either. A positive effect on crumb firmnesswas found in loaves with 35 UX of xylanase per 100 g of flour. Inthis case, the crumb firmness of the loaves with 35 UX reducedin comparison to the control loaves from the third day of storageon, and on the tenth day, this reduction was at a rate of 25%. Jianget al. (2008) reported a reduction in bread crumb firmness by 54%in loaves that had the thermostable xylanase enzyme from T. mari-tima MSB8 added. Waters et al. (2010) also observed a reduction in

Fig. 4. Effect of enzyme cocktail produced by T. aurantiacus CBMAI 756 on crumbfirmness during bread storage at 4 �C. (U/100 g wheat flour).

bread crumb firmness after 5 storage days when they used enzymecocktail with xylanolitic activity produced by thermophlic fungusT. emersonii. The crumb firmness of loaves with 50 UX of xylanaseincreased from the first day on, and was 47% higher than that of thecontrol bread on the last day of storage. These results reinforce thehypothesis of excessive AX hydrolysis when a larger enzyme con-centration was used, which decreases bread quality.

The effect of xylanase activity on amylopectin retrogradation ofwheat starch was evaluated by measuring the increase in the en-thalpy (DH) of the retrogradation endotherm obtained from DSCduring the period of storage. No endotherm was observed at thebeginning of the storage period in any of the loaves studied (Ta-ble 2). This was expected because gelatinized amylopectin retro-gradation is slow. Generally, xylanase at concentrations of 20 andespecially 35 UX per 100 g of flour improved bread quality. Onthe third day of storage, all loaves presented the same DH. Afterthat, the DH in all of the loaves (with exception of those with35 UX of xylanase per 100 g of flour) increased until the fifth dayand remained practically unchanged until the end of storage. Thisincrease was less obvious for loaves with 20 UX of xylanase. On theother hand, the DH in loaves with 35 UX of xylanase did not in-creased after the third day of storage. This result suggests thatthe products that formed through hydrolytic action of xylanaseat this concentration somehow interfered with the reorganizationof the amylopectin and/or with the redistribution of water in thesystem, which reduced the retrogradation of this polymer. Jianget al. (2008) also reported a lower enthalpy of crumbs in loavesthat were added of T. maritima MSB8 xylanase.

From the results, it was possible to observe that an optimal con-centration of the xylanase was 35 UX/100 g of flour. A concentra-tion of 20 UX/100 g of flour was not enough to prevent thecrumb firmness, while a concentration of 50 UX/100 g of flourlikely provoked excessive hydrolysis of the AX, reducing thebread’s overall quality.

This paper investigated the overall effect of the enzyme cocktailproduced by T. aurantiacus CBMAI 756 on wheat bread production.A deeper analysis of the changes in starch, non-starch polysaccha-rides (xylans, b-glucans), and gluten must be performed using awell-characterized purified set of enzyme fractions of this cocktail,and will be the object of future investigations.

4. Conclusion

The thermophilic fungus T aurantiacus CBMAI 756 using corn-cob as a nutrient source on solid-state fermentation produced acocktail of enzymes that had xylanase as major component, whichimproved the quality of bread. Xylanases present in this enzymecocktail broke the b-1,4 glycosidic linkages, especially that of theWU-AX. The WE-AX and partially solubilized WU-AX of HMW ob-tained by the enzyme cocktail action affected the moistureredistribution in the system and contributed to increasing thebread’s specific volume, reducing the bread crumb firmness, and

146 D.S. Oliveira et al. / Food Chemistry 143 (2014) 139–146

becoming amylopectin retrogradation difficult. These resultsclearly present evidence that an enzyme cocktail from T. aurantia-cus positively modifies the wheat dough by improving the proper-ties of the bread obtained and retarding staling and therefore,support the use of this enzyme cocktail as an aid agent in the breadindustry.

Acknowledgement

The authors thank the São Paulo State Research Foundation (FA-PESP) and National Counsel of Technological and Scientific Devel-opment (CNPq) – Brazil, for their financial support.

References

AACC (2000). Approved methods of the American association of cereal chemists. St.Paul, MN, USA: AACC International.

Bergmeyer, H. U., & Bernt, E. (1974). Determination of glucose with oxidase andperoxidase. In H. U. Bergmeyer (Ed.), Methods of enzymatic analysis(pp. 1205–1215). Weinhein: Verlag Chemie.

Courtin, C. M., & Delcour, J. A. (2001). Relative activity endoxylanases towardswater-extractable and unextractable arabinoxylan. Journal of Cereal Science,33(3), 301–312.

Courtin, C. M., & Delcour, J. A. (2002). Arabinoxylans and endoxylanases in wheatflour bread-making. Journal of Cereal Science, 35(3), 225–243.

Courtin, C. M., Roelants, A., & Delcour, J. A. (1999). Fractionation-reconstitutionexperiments provide insight into the role of endoxylanases in bread-making.Journal of Agricultural and Food Chemistry, 47(4), 1870–1877.

Damen, B., Pollet, A., Dornez, E., Broekaert, W. F., Haesendonck, I. V., Trogh, I., et al.(2012). Xylanase-mediated in situ production of arabinoxylan oligosaccharideswith prebiotic potencial in whole meal breads and breads enriched witharabinoxylan rich materials. Food Chemistry, 131(1), 111–118.

Da-Silva, R., Lago, E. S., Merheb, C. W., Macchione, M. M., Park, Y. K., & Gomes, E.(2005). Production of xylanase and CMCase on solid-state fermentation indifferent residues by Thermoascus aurantiacus miehe. Brazilian Journal ofMicrobiology, 36(3), 235–241.

Frederix, S. A., Van, H., Courtin, C. M., & Delcour, J. A. (2004). Water-extractable andwater-unextractable arabinoxylans affect gluten agglomerataion behaviorduring wheat flour gluten-starch separation. Journal of Agricultural and FoodChemistry, 52(26), 7950–7956.

Goesaert, H., Slade, L., Levine, H., & Delcour, J. A. (2009). Amylases and bread firmingan integrated view. Journal of Cereal Science, 50(3), 345–352.

Gray, J. A., & Bemiller, J. N. (2003). Bread staling: Molecular basis and control.Comprehensive Reviews in Food Science and Food Safety, 2(1), 1–21.

Jiang, Z., Bail, A. L., & Wu, A. (2008). Effect of the thermostable xylanase B (XynB)from Thermotoga maritima on the quality of fronzen partially baked bread.Journal of Cereal Science, 47(2), 172–179.

Jiang, Z., Li, X., Yang, S., Li, L., & Tan, S. (2005). Improvement of the breadmakingquality of wheat flour by the hyperthermophilic xylanase B from Thermotogamaritima. Food Research International, 38(1), 37–43.

Keskin, S. Ö., Sumnu, G., & Sahin, S. (2004). Usage of enzymes in a novel bakingprocess. Food/Nahrung, 48(2), 156–160.

Kongbuntad, W., Saenphet, K., Khanongnuch, C., & Lumyong, S. (2006).Determination of toxic effects of crude xylanase derived from Thermoascusaurantiacus SL16W by hematology and blood biochemistry in albino rats.Pakistan Journal of Biological Science, 9(12), 2255–2260.

Kulkarni, N., Shendye, A., & Rao, M. (1999). Molecular and biotechnological aspectsof xylanases. FEMS Microbiology Reviews, 23(4), 411–456.

Laurikainen, T., Härkönen, H., Autio, K., & Poutanen, K. (1998). Effects of enzymes infibre-enriched baking. Journal of the Science of Food and Agriculture, 76(2),239–249.

Martin, M. L., Zeleznak, K. J., & Hoseney, R. C. (1991). A mechanism of bread firming.I. Role of starch swelling. Cereal Chemistry, 68(5), 503–507.

Martinez-Anaya, M. A., & Jiménez, T. (1997). Functionality of enzymes thathydrolyze starch and non-starch polysaccharide in breadmaking. EuropeanFood Research and Technology, 205(3), 209–214.

McCleary, B. V. (1986). Enzymatic modification of plant polysaccharides.International Journal of Biological Macromolecules, 8(6), 349–354.

Miller, G. L. (1959). Use of dinitrosalicylic acid reagent for determination ofreducing sugars. Analytical Chemistry, 31(1), 426–428.

Oliveira, D. S., Meherb-Dini, C., Franco, C. M. L., Gomes, E., & Da-Silva, R. (2010).Production of crude xylanase from Thermoascus aurantiacus CBMAI 756 aimingthe baking process. Journal of Food Science, 75(7), 588–594.

Palma-Fernandez, E. R. D., Gomes, E., & Da-Silva, R. (2002). Purification andcharacterization of two b-glucosidases from thermophilic fungus Thermoascusaurantiacus Miehe. Folia Microbiologica, 47(6), 685–690.

Rantanen, H., Virkki, L., Tuomainen, P., Kabel, M., Schols, H., & Tenkanen (2007).Preparation of arabinoxylobiose from rye xylan using family 10 Aspergillusaculeatus endo-1,4, b-D-xylanase. Carbohydrate Polymers, 68(2), 350–359.

Ribotta, P. D., & Le Bail, A. (2007). Thermo-physical assessment of bread duringstaling. LWT – Food Science and Technology, 40(5), 879–884.

Shah, A. R., Shah, R. K., & Madamwar, D. (2006). Improvement of the quality ofwhole wheat bread by supplementation of xilanase fromAspergillus foetidus.Bioresource Technology, 97(16), 2047–2053.

Subramaniyan, S., & Prema, P. (2002). Biotechnology of microbial xylanases:Enzymology, molecular biology and application. Critical Reviews inBiotechnology, 22(1), 33–64.

Swennen, K., Courtin, C. M., Bruggen, B. V. D., Vandecasteele, C., & Delcour, J. A.(2005). Ultrafiltration and ethanol precipitation for isolation ofarabinoxylooligosaccharides with different structures. Carbohydrate Polymers,62(3), 283–292.

Verjans, P., Dornez, E., Delcour, J. A., & Courtin, C. M. (2010). Selectivity for water-unextractable arabinoxylan and inhibition sensitivity govern the strong breadimproving potential of an acidophilic GH11 Aureobasidium pullulans xylanase.Food Chemistry, 123(2), 331–337.

Waters, D. M., Murray, P. G., Ryan, L. A., Arendt, E. K., & Tuohy, M. G. (2010).Talaromyces emersonii thermostable enzyme systems and their applications inwheat baking systems. Journal of Agricultural and Food Chemistry, 58(12),7415–7422.

Waters, D. M., Ryan, L. A. M., Murray, P. G., Arendt, E. K., & Tuohy, M. G. (2011).Characterisation of a Talaromyces emersonii thermostable enzyme cocktail withapplications in wheat dough rheology. Enzyme and Microbial Technology, 49(2),229–236.