aem00035-0041

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, June 1993, p. 1725-17300099-2240/93/061725-06$02.00/0Copyright ©) 1993, American Society for Microbiology

Purification and Characterization of a Thermostable Xylanasefrom Bacillus stearothermophilus T-6

ALEXANDER KHASIN, IRIS ALCHANATI, AND YUVAL SHOHAM*

Department ofFood Engineering and Biotechnology, The Technion, Technion City, Haifa 32000, Israel

Received 28 December 1992/Accepted 14 March 1993

BaciUlus stearothermophilus T-6 produces an extracellular xylanase that was shown to optimally bleach pulpat pH 9 and 65°C. The enzyme was purified and concentrated in a single adsorption step onto a cationexchanger and is made of a single polypeptide with an apparent Mr of 43,000 (determined by sodium dodecylsulfate-polyacrylamide gel electrophoresis). Xylanase T-6 is an endoxylanase that completely degrades xylan toxylose and xylobiose. The pls of the purified protein were 9 and 7 under native and denaturing conditions,respectively. The optimum activity was at pH 6.5; however, 60% of the activity was still retained at pH 10. At65°C and pH 7, the enzyme was stable for more than 10 h; at 65°C and pH 9, the half-life of the enzyme was

approximately 6 h. Kinetic experiments at 55°C gave V.. and Km values of 288 U/mg and 1.63 mg/ml,respectively. The enzyme had no apparent requirement for cofactors, and its activity was strongly inhibited byZn2+, Cd2 , and Hg2'. Xylan completely protected the protein from inactivation by N-bromosuccinimide. TheN-terminal sequence ofthe first 45 amino acids of the enzyme showed high homology with the N-terminal regionof xylanase A from the alkalophilic Bacillus sp. strain C-125.

Xylanases (1,4-P-D-xylan xylanohydrolase; EC 3.2.1.8)are hemicellulases that hydrolyze xylan, which is a majorconstituent of the hemicellulose complex (6). Xylan is com-posed of ,3-1,4-linked xylopyranose units with branchescontaining L-arabinofuranosyl and glucopyranosyl residues.Biotechnological uses and potential applications of xyla-nases include bioconversion of lignocellulose material tofermentative products, clarification of juices, and improve-ments of the consistency of beer and the digestibility ofanimal feedstock (45). One of the major potential applica-tions of xylanases involves the pulp and paper industry (44).In the process of making paper pulp, lignin is removed bycooking wood chips at a high temperature and a basic pH(the Kraft process). The residual lignin that remains on thepulp is dark in color because it has been extensively oxidizedand modified in the cooking process. To obtain high-qualitywhite paper, all of the lignin must be removed; this is donetraditionally with chlorine-based chemicals (bleaching) (38).This bleaching process, although highly effective, produceschlorinated organics and is polluting (40). Today, the pulpand paper manufacturers are actively seeking new bleachingprocedures that will reduce or even eliminate the need forchlorine bleaching (36). Several years ago Viikari et al.demonstrated that hemicellulases can be used to enhancedelignification and bleaching of unbleached pulp (43). Themajor effect of the enzymes is due to the hydrolysis ofreprecipitated and readsorbed xylan and xylan-lignin com-

plexes that are separated during the cooking process. Acomparison of different hemicellulases indicated that endo-,B-xylanases have a major impact on delignification, even insoftwood pulp, of which mannan is a major component.Indeed, full-scale mill trials of enzyme prebleaching are

already under way (11, 24, 43). Most of the hemicellulasesstudied to date are active at a neutral or acidic pH, and theiroptimal activity temperature is below 45°C. Hemicellulasesthat are active at higher temperatures and pHs are of great

* Corresponding author. Electronic mail address: forO610@technion.

potential since they can be introduced more freely in thedifferent stages of the bleaching line without the need ofcostly changes in temperature and pH. Recently, we haveisolated thermostable alkaline-tolerant xylanases that canbleach pulp optimally at pH 9 and 65°C (16, 35, 37). In thisarticle, we report the purification and characterization of oneof these enzymes, xylanase T-6.

MATERIALS AND METHODS

Organisms and culture conditions. Bacillus stearothenno-philus T-6 was isolated by using an enrichment procedure forbacteria capable of producing extracellular thermostablexylanases (37). Strain T-6 was identified as B. stearothermo-philus by the National Collection of Industrial and MarineBacteria (Torry Research Station, Aberdeen, Scotland) andwas designated NCIMB 40221. For xylanase production,fermentation of the organism was carried out in a 5-liter NewBrunswick Microferm fermentor at 60°C and 600 rpm andwith aeration of 8 to 10 liters/min for 12 to 18 h. Growthmedium, XMP, contained the following: vitamin assayCasamino Acids (Difco), 4.0 g/liter; yeast extract, 0.2 g/liter;MgSO4, 0.01 g/liter; (NH4)2SO4, 2.0 g/liter; K2HPO4, 0.46g/liter; KH2PO4, 0.1 g/liter; MOPS (morpholinepropane-sulfonic acid), 10.4 g/liter; D-xylose (autoclaved separately),5.0 g/liter; and trace element solution, 1 ml/liter. The traceelement solution contained the following (in grams per liter):CaCl2 2H20, 0.39; CuSO4 5H20, 0.62; FeSO4. 7H20,0.60; MnSO4, 0.59; ZnSO4- 7H20, 0.42; CoCl2 6H20, 0.79;and Na2MoO4, 0.70. The solution was kept at pH 2 andadded after sterilization of the medium.Enzyme purification. Enzyme purification was carried out

by a single batch adsorption step onto a cation exchanger(usually 3 to 5% SE-52; Whatman Ltd., Maidstone, En-gland). The ionic strength of the broth or cell-free broth(cells were removed by centrifugation at 10,000 x g and 4°Cfor 10 min) was adjusted to give specific conductivity of lessthan 5 mS/cm2 (below 50 mM KCl equivalent), and theadsorption step was allowed to proceed for at least 30 min.Elution was carried out with 1 M KCl after the cation

1725

Vol. 59, No. 6

APPL. ENVIRON. MICROBIOL.

exchanger had been washed with low-strength buffer (10 mMphosphate buffer [pH 7]).

Xylanase assay. Appropriately diluted enzyme (250 ,ul) wasmixed with 250 ,ul of 2% oat spelts xylan (Sigma ChemicalCo., St. Louis, Mo.) and 500 ,ul of 0.1 M phosphate buffer(pH 7.0). (Xylan solution was prepared by sonicating a 2%xylan solution for 3 min with an Ultrasonic W375, sonicatorat output 7.) Four aliquots of 0.1 ml were taken from thismixture and placed in four 9-ml glass tubes. Two tubesserved as time-zero controls and were kept at room temper-ature or on ice; the two other tubes were incubated at 55 or65°C for 10 or 15 min. The reaction was terminated byplacing the tubes in a water bath at room temperature. Thereducing sugar content in the tubes was determined by thedinitrosalicylic acid (DNS) method (27), with D-xylose as astandard (0.004% xylose was added to the DNS reagent justbefore the color reaction). One unit of xylanase activity wasdefined as the amount of enzyme which produces 1 ,umol ofxylose equivalent per min.

Effects of pH and temperature on xylanase activity. Theenzymatic reactions were carried out for 5 min at 55°C infour different buffers (50 mM): citric acid-Na2HPO4 (pH 5 to6), phosphate buffer (pH 6 to 8), boric acid-NaOH buffer (pH8 to 9.5), and phosphate-NaOH buffer (pH 9.5 to 11). Theactual pH in the assay mixture was determined at thereaction temperature. The effect of temperature on thereaction rate was determined by performing the standardreaction for 5 min at a temperature range of 55 to 80°C.

Thermostability. Eppendorf tubes (1.5 ml) containing 0.2ml of purified enzyme solution (40 U/ml in 10 mM phosphatebuffer [pH 7.0]) were incubated at 65, 70, and 75°C. Atvarious times, the tubes were removed and placed at -20°C.The residual enzymatic activity in each tube was determinedby the standard assay.

Effect of metals. Various salts (at 0.1, 1.0, and 10 mM)were added to the standard enzymatic reaction mixtures (toavoid the formation of insoluble phosphates, 50 mM succinicacid buffer [pH 7.0] was used instead of the phosphatebuffer).

Protein content. Protein content was determined by themethod of Bradford (5) by using the Bio-Rad protein assay(Bio-Rad Laboratories, Richmond, Calif.) with bovine albu-min fraction V (Sigma) as a standard.

Isoelectric point. Isoelectric focusing of the native proteinwas performed by the procedure described by Guilian et al.(12) with ampholytes of pH 3.5 to 10 (Sigma). Isoelectricfocusing under denaturing conditions was performed by theprocedure of O'Farrell et al. (29). Chromatofocusing wasperformed on a fast protein liquid chromatography (FPLC)Mono P column (Pharmacia, Uppsala, Sweden) with Poly-buffer 96 (Pharmacia) as an elution buffer.

Molecular weight. The Mr of the purified enzyme wasestimated by sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE) (25) and by gel filtration onSephadex G-75 and Superose 12 FPLC columns (Pharma-cia).

Thin-layer chromatography. Xylan digestion productswere analyzed on a silica gel chromatogram. Samples ofxylan digests (-20 ,ug) were applied to silica gel-coatedaluminum foils (Merck 60 254; E. Merck AG, Darmstadt,Germany). The running solvent consisted of n-butanol-acetone-water (4:5:1 [vol/vol/vol]). Xylose and xylobiose(Sigma) were used as standards. After the run (2 h), thechromatogram was sprayed with a mixture of 0.5 ml ofanisaldehyde, 50 ml of acetic acid, and 1 ml of H2SO4 and

TABLE 1. Single-step purification of B. stearothermophilusT-6 xylanase

Vol Total Sp act YieldFraction (MI) activity (U/mg) (%Fraction (Vml) (U)

Crude broth 100 145 7.2 100Dialyzed broth 100 130 7.2 90CM-52 adsorption and 29 66.7 280 46KCI elution (1 M)

heated at 105°C for 10 min. Pentoses gave green to grayspots.Enzyme modifications. Enzyme (1 p,g) was added to tubes

containing 0.5 ml of 50 mM succinic acid buffer (pH 6.0) withdifferent amounts of xylan (0 to 5 mg). The modifier,N-bromosuccinimide at 4 ,uM orp-hydroxymercuribenzoateat 1 mM (both from Sigma), was added to the tubes, and themixtures were incubated for 10 min at 25°C. Residual activ-ity of the enzyme was determined in the standard assay byadding 0.5% xylan to the reaction tubes.Mode of action. To test whether xylanase T-6 is an

endoxylanase, viscosity and reducing sugar content weredetermined in parallel. The enzymatic reaction, carried outin 50 mM phosphate buffer (pH 7.0) containing xylan (0.4%),was performed directly in an Ostwald capillary viscometer(Volac size 50) while the viscosity was being measured (thebuffer value was 184 s at 65°C).Amino acid analyses. Total amino acid analysis was done

on a Dionex amino acid analyzer, by using ion-exchangechromatography with ninhydrin. In addition, the number ofthiols in the protein was estimated by the Ellman assay afterthe enzyme was reduced with dithiothreitol (8). The N-ter-minal amino acid sequence of xylanase T-6 was determinedwith an Applied Biosystems model 475A gas-phase se-quencer.

RESULTS AND DISCUSSION

Enzyme production and purification. B. stearothermo-philus T-6 was isolated by using an enrichment and screeningprocedure for extracellular xylanases that are thermostableand alkaline tolerant (37). Extracellular xylanase activitywas found to be present in the growth media of B. stearo-thermophilus T-6 grown in the presence of xylose. In atypical fermentation on XMP medium, cell turbidity reached500 to 600 Klett units and the extracellular xylanase activitywas about 2 U/ml. The activity in the broth could beconcentrated either by ammonium sulfate or acetone precip-itation. The most efficient way of concentrating the enzyme,with no loss of activity, was with dialysis tubing placedagainst solid polyethylene glycol 20000. The enzyme couldalso be concentrated and purified by adsorbing it onto acation exchanger such as CM-11, CM-52, or SE-52 (What-man). It was possible to efficiently adsorb the enzymedirectly from the fermentation broth, provided that the ionicstrength of the broth was sufficiently low (below 50 mM KCIequivalent). Reduction of the ionic strength was achieved byeither dialyzing the broth against a low-strength buffer or,alternatively, by omitting the MOPS buffer from the mediumand controlling the pH externally (i.e., via a pH controller).Typical results of this single-step purification scheme areshown in Table 1. The total yield was 46%, with a purifica-tion factor of about 42. The enzyme obtained in the KCIeluent was more than 99% pure as judged by FPLC gel

1726 KHASIN ET AL.

XYLANASE FROM B. STEAROTHERMOPHILUS T-6 1727

, s

*,t*00-0Y-~

C

.

v.

::-4

MW

106,500

" 80,000

0

>o

o

0-

4.*-

._i

100

80

60

40

20

o

i.

49,500

- 32,500

FIG. 1. SDS-PAGE of purified and crude xylanase T-6. Lanes:A, concentrated crude broth; B, cation exchanger-purified xylanase;C, FPLC-purified xylanase after a cation exchanger purification.

filtration, N-terminal analysis, and SDS-PAGE (Fig. 1). Ahigh degree of purification (40- to 50-fold), with a yield of 40to 50%, was achieved repeatedly in both small- (5-liter) andlarge-scale (1,000-liter) purification experiments. This simpleand effective batch adsorption procedure was possible be-cause of (i) the high selectivity of the cation exchangeradsorption step and (ii) the high partition coefficient ofxylanase T-6 to the adsorbent (37). Purification yields forother xylanases usually lie in the range of 1 to 30% (1, 3, 10,15, 30, 46). A similar single-step purification procedure wasreported for a- and ,-galactosidases from Aspergillus niger(7).

Molecular weight determination. The molecular weight ofthe purified xylanase T-6 was estimated by both SDS-PAGEand by gel filtration on Sephadex G-75 and FPLC Superose12 HR 10/30 columns (Pharmacia). The SDS-PAGE deter-mination gave a molecular weight of 43,000 (Fig. 1). How-ever, estimation of the molecular weight by gel filtration onSephadex G-75 gave varied results, strongly dependent onthe ionic strength of the elution buffer. The retention coef-ficient of the enzyme decreased with the increase of thebuffer ionic strength (in the range of 20 to 100 mM phosphatebuffer [pH 7]), suggesting that the enzyme interacts withSephadex. At buffer concentrations of 100 mM or higher, theenzyme was eluted as a sharp peak corresponding to amolecular weight of 31,000. Gel filtration on Superose 12 (theelution buffer was 100 mM phosphate buffer [pH 7]-100 mMNaCl-0.02% NaN3) gave retention coefficients of 1.92, 1.76,and 2.05 for xylanase T-6, bovine serum albumin (molecularweight, 66,000), and carbonic anhydrase (molecular weight,29,000), respectively, suggesting a molecular weight of41,900 for xylanase T-6. The enzyme probably consists of asingle polypeptide chain, since the molecular weight estima-tions by SDS-PAGE (43,000) and FPLC gel filtration (41,900)are similar.

pl. The isoelectric point of xylanase T-6 was determined

4 5 6 7 8 9 10 11

pHFIG. 2. Effect of pH on the initial reaction rates of xylanase T-6.

Buffers (50 mM) used include citric acid (0), phosphate (0), boricacid (-), and phosphate-NaOH (O).

under native and denaturing conditions. The native proteinexhibited a pI of 9.0 on isoelectric focusing gels, and, inagreement with this result, the protein was eluted at pH 9.0from a chromatofocusing column. Under denaturing condi-tions, the pI of the enzyme was 7.0. Under both conditions,the purified protein gave three minor bands on the isoelectricfocusing gels. These bands probably reflect small changes ofcharged groups on the enzyme. The differences in the pl ofthe minor components were +0.15 pH unit. The relativelyhigh pI of the native protein (9.0) explains its positive chargeat neutral pHs and the fact that it can be selectively adsorbedonto a cation exchanger from the crude supernatant.

Kinetics of xylan degradation. Xylanase T-6 hydrolyzes oatspelts xylan to release reducing sugars. At 55°C, the releaseof reducing sugar was linear with time (for at least 20 min)and proportional to enzyme concentration (for formation ofup to 1.5 ,umol of xylose equivalent in the reaction tubes).Kinetic experiments at 55°C with different xylan concentra-tions gave a Vm, of 288 U/mg and a Km of 1.63 mg/ml. Theturnover number of the enzyme was 12,382 (moles of reduc-ing ends released per mole of enzyme per minute), calculatedfrom its Vm. and the estimated molecular weight of 43,000.

Effect of pH. The pH range at which xylanase T-6 is activewas determined in four different buffers covering the rangebetween pH 5.0 and 11.0. The enzyme was most active in theneutral pH range, between pH 6.5 and 7.0, but retained 60%of its activity at pH 10 (Fig. 2). At pHs below 4.5, theenzyme tended to precipitate in our assay conditions. Thealkaline tolerance of xylanase T-6 is crucial, considering itspotential application in bleaching pulp at high pH values.Indeed, the enzyme was shown to bleach pulp optimally atpH 9.0 and 65°C (16). Most xylanases known today areactive at acidic (20, 39) or neutral pHs (2, 3, 26, 30).Recently, however, several alkaline-tolerant xylanases werecharacterized (1, 9, 18, 19, 31-33, 41).

Reaction rate at different temperatures. Initial reactionrates were determined at temperatures between 45 and 85°C.The highest initial reaction rate was obtained at 75°C. Therelative activities at the different temperatures were 28, 51,70, 81, 90, 100, 74, and 40% at 45, 55, 60, 65, 70, 75, 80, and85°C, respectively. The Arrhenius plot-calculated activation

I . I . I . I . I .

0

0 0

0

: 0

*"|.|.|.l~~~~~~~~~~~m

VOL. 59, 1993

APPL. ENVIRON. MICROBIOL.

' 80>k

,60

40.

* 20

00 2 4 6 8 10 12

Time (hours)FIG. 3. Thermostability of xylanase T-6 at pH 7.0. Residual

xylanase activity was monitored at various times after incubation at65°C (0), 70°C (0), and 75°C (-).

energy, 9.3 kcal/mol (ca. 39 kJ/mol), is in the range that ischaracteristic of typical enzymatic reactions.

Thermostability. Thermoinactivation experiments for xy-lanase T-6 were performed by incubating the enzyme solu-tion at different temperatures and determining the residualactivity at various times. The thermoinactivation of theenzyme at pH 7.0 is shown in Fig. 3. Exposure of theenzyme for more than 10 h at 65°C did not affect the activity.At 70 and 75°C, the half-lives of the enzyme were about 14.5h and 20 min, respectively. At pH 9.0 and 65°C, the half-lifeof the enzyme was about 6 h. The activation energy of thethermoinactivation mechanism can be estimated from thekinetics of thermoinactivation at pH 7 and is equal to 119kcal/mol (ca. 498 kJ/mol). This value, which is uncharacter-istic of covalent reactions, suggests that thermoinactivationis controlled by a monomolecular conformational process(unfolding of the native protein) (22, 23).

Effect of metal ions. The effect of different metals on theactivity of xylanase T-6 is shown in Table 2. The metal ions

TABLE 2. Effect of metal ions on xylanase T-6 activity

Xylanase activity (%)Chemical

10 mM 1 mM 0.1 mM

None 100 100 100LiCl 101 95NaCl 110 94KCl 110 101MgCl2 109 101CaSO4 88 102BaSO4 90 96FeSO4 63 104Ni(NO3)2 51 98MnCl2 62 103CoCl2 78 104CuS04 48 101Ag(NO3)2 67 100ZnCl2 38 98CdCl2 47 90HgCl2 17 70A1(NO3)3 54 91EDTA 100

0

0)-._

ncon5

iE

C0

0*._

crw

005.X

50 100Time (min)

FIG. 4. Drop in viscosity and formation of reducing powerduring the degradation of xylan by xylanase T-6. The reaction wascarried out on 0.4% xylan at 65°C. Symbols: 0, relative viscosity;0, xylose equivalent.

Fe2+, Ni2+, Mn2+, Co2+, Cu2+, Ag2+, and Al` showedsome inhibition of the activity, while the group Ilb metals,Zn2+, Cd2+, and Hg2+, gave the strongest inhibition. Thegroup Ilb metals exhibit high affinity for reactive groups. Forexample, the affinity of Hg toward reactive groups is SH >CONH2> NH2> COOH> PO4 (42). Monovalent cationssuch as Li+, Na+, and K+ had small stimulating effects onthe activity. The addition of EDTA did not affect theactivity, suggesting that no metals are needed for the enzy-matic reaction.Mode of action. To test whether xylanase T-6 is an

endoxylanase, the decrease of viscosity and formation ofreducing sugar were determined in parallel. The action ofxylanase T-6 on xylan was characterized by a rapid reduc-tion in the viscosity of the substrate and a relatively slowincrease in the concentration of reducing sugar (Fig. 4). Theincrease in reducing sugar from 0.4 to 1.7 ,umol of xyloseequivalent per ml (an average of two breakages per mole-cule) results in a 25% decrease in viscosity. On the basis ofthe Einstein equation, which relates molecular weight toviscosity, it is clear that such reduction in viscosity canresult only from an endo-cleavage of the molecules. Toexamine the extent to which xylanase T-6 can degrade xylan,a complete digestion experiment was performed. The stan-dard enzymatic assay reaction (0.5% xylan) continued for 20h at 65°C, and samples were taken for reducing sugaranalysis and sugar composition analysis on thin-layer chro-matography. The enzyme solution (10 ,ul of a 40-U/mlconcentration) was added periodically to the reaction mix-ture to facilitate the degradation of the polymer. The com-plete digestion of oat spelts xylan produced about 25 ,umol ofxylose equivalent per ml, compared with the theoreticalmaximum value of 37.8 ,umol/ml. The average size of the endproduct is about 1.5 xylopyranose units, possibly corre-sponding to a mixture of xylose and xylobiose. Thin-layerchromatography of the end products also indicated thatxylose and xylobiose were the only main products after the20-h digestion (data not shown). Nanmori et al. (28) charac-terized a xylanase from B. stearothermophilus 21. Theaction of this enzyme together with 3-xylosidase on xylangave only xylose, suggesting that xylanase 21 and xylanaseT-6 have similar modes of action. The two enzymes, how-ever, have different molecular masses and isoelectric points.

Substrate specificity. The activity of xylanase T-6 was

1728 KHASIN ET AL.

XYLANASE FROM B. STEAROTHERMOPHILUS T-6 1729

TABLE 3. Xylanase T-6 substrate specificity

Substrate" Related enzyme (U/ag)bXylan (oat spelts) 1,4-D-Xylanase 288Guar gum D-Mannanase <0.01Locust bean gum D-Mannanase <0.01Arabinogalactan D-Galactanase <0.01CMC Endo-1,4-p-glucanase 0.03PNP-P-D-cellobioside Exo-1,4-P-glucanase 3.10PNP-P-D-xylopyranoside P-D-Xylosidase 0.20PN1-ct-L-arabinopyranoside a-L-Arabinopyranosidase 0.04PNP-a-L-arabinofuranoside a-L-Arabinofuranosidase 0.03PNP-P-D-galactopyranoside P-D-Galactosidase 0.01PNP-a-D-glucopyranoside a-D-Glucosidase 0.02PNP-P-D-glucopyranoside ,-D-Glucosidase 0.02

a PNP, p-nitrophenyl; CMC, carboxymethyl cellulose.b Reactions were carried out at 55'C with a highly concentrated and purified

enzyme. A unit of activity was defined as the amount of enzyme whichreleased 1 ,umol of either reducing sugar equivalent orp-nitrophenol per min.With the following substrates the activity was less than 0.01 U/mg: PNP-13-L-arabinopyranoside, PNP-a-L-fucopyranoside, PNP- -L-fucopyranoside,PNP-a-L-rhamnopyranoside, PNP-3-D-mannopyranoside, PNP-a-D-manno-pyranoside.

tested on several cellulose- and hemicellulose-related sub-strates (Table 3). The enzyme can be classified as a type IIaendoxylanase, which cannot cleave L-arabinosyl branchpoints but does cleave xylooligosaccharides as short asxylotriose and produces mainly xylose and xylobiose as finalproducts (34). The very low activity on carboxymethylcellulose is an advantage, considering the potential applica-tion of the enzyme in biobleaching pulp for high-qualitypaper.Amino acid analyses. The amino acid composition of

xylanase T-6 is given in Table 4. The enzyme is rich in lysineand probably has no cysteines. No thiol groups were de-tected with the Ellman assay. The first 45 amino acids fromthe N terminus of xylanase T-6 were sequenced. Thissequence was analyzed against the GenBank data banklibraries with TFasta (GCG software package; GeneticsComputer Group, University of Wisconsin, Madison) andshowed significant homology (45.5% on 33 amino acidsoverlap) only with the N-terminal region of xylanase A fromthe alkalophilic Bacillus sp. strain C-125 (13, 14, 17, 18) (Fig.5). The two enzymes have similar molecular weights, exhibit

TABLE 4. Amino acid composition of xylanase T-6

Concn Residues/(nmol) molecule

Aspartic acid (Asp + Asn) 4.71 58Threonine 0.85 12Serine 0.66 10Glutamic acid (Glu + Gln) 3.64 44Proline 1.95 24Glycine 1.64 20Alanine 2.44 30Valine 2.25 28Methionine 0.15 2Isoleucine 2.06 26Leucine 1.03 14Tyrosine 1.61 22Phenylalanine 1.27 16Lysine 3.12 38Histidine 0.56 6Arginine 0.90 12

1 Xylanase T-6Lys Asn Ala Asp Ser Tyr Ala Lys Lys Pro flia Ile Ser Ala LeuVal Phe Gly Glu Asn Glu Lys Arg Asn Asp Glan Pro Phe Ala Trp10 Xylanase A

16Aan AJA BroQ Gln Leu AS; Lin Arg Tyr Ly- Asn Glu Phe Thr IleGln VYal Ala Ser Leu Ser Glu Arg Tyr Gln Glu GUn Phe Asp Ile26

31Gly Ala Ala Val Glu Pro Tyr Gln Leu Gln Asn Glu Lys Asp ValGly Ala Ala Val Glu Pro Tyr Gln Leu Glu Gly Arg Gln Ala Gln41

FIG. 5. Alignment of the N-terminal sequence of xylanase T-6with the N-terminal sequence of xylanase A from the alkalophilicBacillus sp. strain C-125. Identical amino acids are in boldface type;conservative amino acids replacements are underlined.

broad pH activity curves, and are both active at an alkalinepH. However, xylanase T-6 is more thermostable and canhydrolyze xylotriose to xylose and xylobiose, and aftercomplete hydrolysis of xylan, its end products are onlyxylose and xylobiose. It will be interesting to compare thecomplete amino acid sequence of the two enzymes and toidentify the regions responsible for their exceptional alkalineand heat tolerance properties.

Chemical modifications of xylanase T-6. Tryptophan andcysteine were shown to be involved in the active site ofdifferent xylanases (4, 21, 26). To test whether these residuesare present in the active or binding site of xylanase T-6, weexamined the ability of xylan (the substrate) to protect theenzyme from tryptophan or cysteine modifiers. Xylanase T-6(2 p,g/ml) was completely inhibited by 4 ,uM ofN-bromosuc-cinimide (a tryptophan modifier); however, xylan at a con-centration as low as 1 mg/ml gave over 95% protectionagainst this reagent. Only about 15% inhibition was detectedafter treatment with 1 mM ofp-hydroxymercuribenzoate (acysteine modifier), indicating again that cysteine is notpresent in the protein. The slight inhibition is probably dueto the reaction of this modifier with other residues.

ACKNOWLEDGMENTS

This work was supported by a grant from Biovik AB and KorsnasAB. Technical support was provided by the Technion-Otto Meyer-hof Biotechnological Laboratories.We thank Orit Gat, Eugene Rosenberg, Raphael Lamed, and the

Board Paper Pulp-Development group at Korsnfis for helpful com-ments on the manuscript.

REFERENCES1. Akiba, T., and K. Horikoshi. 1988. Xylanases of alkalophilic

thermophilic Bacillus. Methods Enzymol. 160:655-659.2. Berenger, J., F. Chantal, J. Bigliardi, and N. Creuset. 1985.

Production, purification and properties of thermostable xyla-nase from Clostridium stercorarium. Can. J. Microbiol. 31:635-643.

3. Bernier, R., M. Desrochers, L. Jurasek, and M. G. Paice. 1983.Isolation and characterization of a xylanase from Bacillussubtilis. Appl. Environ. Microbiol. 46:511-514.

4. Biswas, S. R., S. C. Jana, A. K. Mishra, and G. Nanda. 1990.Production, purification and characterization of xylanase from ahyperxylanolytic mutant of Aspergillus ochraceus. Biotechnol.Bioeng. 35:244-251.

5. Bradford, M. M. 1976. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72:248-254.

6. Browning, B. L. 1963. The composition and chemical reactionsof wood, p. 58-101. In B. L. Browning (ed.), The chemistry ofwood. John Wiley & Sons, Inc. New York.

7. Christakopoulos, P., B. J. Marcris, and D. Kekos. 1990. Excep-

VOL. 59, 1993

APPL. ENVIRON. MICROBIOL.

tionally thermostable a- and P-galactosidases from Aspergillusniger separated in one step. Proc. Biochem. Int. 210-212.

8. Creighton, T. E. 1989. Disulphide bonds between cysteineresidues, p. 157. In T. E. Creighton (ed.), Protein structure, apractical approach. IRL Press, Oxford.

9. Dey, D., J. Hinge, A. Shendye, and M. Rao. 1991. Purificationand properties of extracellular endoxylanases from alkalophilicthermophilic Bacillus sp. Can. J. Microbiol. 38:436-442.

10. Fournier, R. A., M. M. Fredrick, J. R. Fredrick, and P. J. Reilly.1985. Purification and characterization of endo-xylanase fromAspergillus niger. Biotechnol. Bioeng. 27:539-546.

11. Grant, R. 1991. First mill-scale trials get underway. Pulp PaperInt. 33:61-63.

12. Guilian, G. G., R. L. Moss, and M. Greaser. 1984. Analyticalisoelectric focusing using a high voltage vertical slab polyacryl-amide gel system. Anal. Biochem. 142:421-436.

13. Hamamoto, T., H. Honda, T. Kudo, and K. Horikoshi. 1987.Nucleotide sequence of the xylanase A gene of alkalophilicBacillus sp. strain C-125. Agric. Biol. Chem. 51:953-955.

14. Hamamoto, T., and K. Horikoshi. 1987. Alkalophilic Bacillusxylanase A, a secretable protein through outer membrane ofEscherichia coli. Agric. Biol. Chem. 51:3133-3135.

15. Hashimoto, S., T. Muramatsu, and M. Funatsu. 1971. Studies onxylanase from Trichoderma viride. Agric. Biol. Chem. 35:501-508.

16. Hogman, S., H. Joves, E. Rosenberg, and Y. Shoham. 1992.Bleachability improvement of softwood kraft pulp through treat-ment with an alkali- and thermostable xylanase, p. 107-113. InM. Kuwahara and M. Shimada (ed.), Biotechnology in pulp andpaper industry. Uni Publishers, Tokyo.

17. Honda, H., T. Kudo, and K. Horikoshi. 1985. Molecular cloningand expression of the xylanase gene of alkalophilic Bacillus sp.strain C-125 in Escherichia coli. J. Bacteriol. 161:784-785.

18. Honda, H., T. Kudo, Y. Ikura, and K. Horikoshi. 1985. 'rwotypes of xylanases of alkalophilic Bacillus sp. no. C-125. Can. J.Microbiol. 31:538-542.

19. Horikoshi, K., and Y. Atsukawa. 1973. Xylanase produced byalkalophilic Bacillus no. C-59-2. Agric. Biol. Chem. 37:2093-2103.

20. John, M., B. Schmidt, and J. Schmidt. 1979. Purification andsome properties of five endo-1,4-p-D-xylanases and a P-D-xylosidase produced by a strain of Aspergillus niger. Can. J.Biochem. 57:125-134.

21. Keskar, S. S., C. Srinivasan, and V. Deshphande. 1989. Chem-ical modification of a xylanase from thermotolerant Streptomy-ces. Biochem. J. 261:49-55.

22. Klibanov, A. M. 1983. Stabilization of enzymes against thermalinactivation. Adv. Appl. Microbiol. 29:1-24.

23. Klibanov, A. M., and D. B. Volkin. 1989. Minimizing proteininactivation, p. 1-24. In T. E. Creighton (ed.), Protein function,a practical approach. IRL Press, Oxford.

24. Koponen, R. 1991. Enzyme systems prove their potential. PulpPaper Int. 33:81-83.

25. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T-4. Nature (London)227:680-685.

26. Marui, M., K. Nacanishi, and T. Yasui. 1985. Purification andproperties of three types of xylanases induced by methyl-p-xyloside from Streptomyces sp. Agric. Biol. Chem. 49:3399-3407.

27. Miller, G. L. 1959. Use of dinitrosalicylic acid reagent for

determination of reducing sugars. Anal. Chem. 31:426-428.28. Nanmori, T., T. Watanabe, R. Shinke, A. Kohno, and Y.

Kawamura. 1990. Purification and properties of thermostablexylanase and 3-xylosidase produced by a newly isolated Bacil-lus stearothermophilus strain. J. Bacteriol. 172:6669-6672.

29. O'Farrell, P. Z., H. M. Goodman, and P. H. O'Farrell. 1977.High resolution two-dimensional electrophoresis of basic aswell as acidic proteins. Cell 12:1133-1142.

30. Okada, H., and A. Shinmyo. 1988. Xylanase from Bacilluspumilus. Methods Enzymol. 160:632-637.

31. Okazaki, W., T. Akiba, K. Horikoshi, and R. Akahoshi. 1984.Production and properties of two types of xylanases fromalkalophilic thermophilic Bacillus spp. Appl. Microbiol. Bio-technol. 19:335-340.

32. Okazaki, W., T. Akiba, K. Horikoshi, and R. Akahoshi. 1985.Purification and characterization of xylanases from alkalophilicthermophilic Bacillus spp. Agri. Biol. Chem. 49:2033-2039.

33. Ratto, M., K. Poutanen, and L. Viikari. 1992. Production ofxylanolytic enzymes by an alkalitolerant Bacillus circulansstrain. Appl. Microbiol. Biotechnol. 37:470-473.

34. Reilly, P. J. 1981. Xylanases: structure and functions, p. 111-129. In A. E. Hollaender and R. Robson (ed.), Trends in thebiology of fermentations for fuels and chemicals. Plenum Pub-lishing Corp., New York.

35. Rosenberg, E., and Y. Shoham. August 1991. A preparationexhibiting enzymatic delignification activity, a method of pro-ducing the same, and applications thereof. Sweden patent no.465 320.

36. Samadni, G. 1991. Pulp bleaching-the race for safer methods.Chem. Eng. (Int. Ed.) 98:37-43.

37. Shoham, Y., Z. Schwartz, A. Khasin, 0. Gat, Z. Zosim, and E.Rosenberg. Delignification of wood pulp by a thermostablexylanase from Bacillus stearothermophilus strain T-6. Biodeg-radation, in press.

38. Singh, P. S. 1979. Principles of pulp bleaching, p. 15-28. In P. S.Singh (ed.), The bleaching of pulp, 3rd ed. Tappi Press, Atlanta.

39. Tavobilov, I. M., I. V. Gorbacheva, N. A. Rodionova, and A. M.Bezborodov. 1981. Purification of endo-1-4-3-xylanase from thefungus Aspergillus niger strain 15. Appl. Biochem. Microbiol.17:320-324.

40. Trubacek, I., and A. Wiley. 1979. Bleaching and pollution, p.423-461. In P. S. Singh (ed.), The bleaching of pulp, 3rd ed.Tappi Press, Atlanta.

41. Tsujibo, H., T. Sakamoto, N. Nishino, T. Hasegawa, and Y.Inamori. 1990. Purification and properties of three types ofxylanases produced by an alkalophilic actinomycete. J. Appl.Bacteriol. 69:398-405.

42. Venugopal, B., and T. D. Luckey. 1978. Metal toxicity inmammals, p. 68-89. Plenum Press, New York.

43. Viikari, L., A. Kantelinen, M. Ratto, and J. Sundquist. 1991.Enzymes in pulp and paper processing, p. 12-21. ACS Symp.Ser. 460 (Enzymes Biomass Conversion). American ChemicalSociety, Washington, D.C.

44. Viikari, L., J. Sundquist, and J. Kettunen. 1991. Xylanaseenzymes promote pulp bleaching. Paperi ja Puu - Paper andTimber 73:384-389.

45. Wong, K. Y., L. U. L. Tan, and J. N. Saddler. 1988. Multiplicityof 0-1,4-xylanase in microorganisms: functions and applica-tions. Microbiol. Rev. 52:305-317.

46. Yasui, T., M. Marui, I. Kusakabe, and K. Nakanishi. 1988.Xylanases of Streptomyces. Methods Enzymol. 160:648-654.

1730 KHASIN ET AL.