Vol. 171, No. 6

Cloning of a Thermomonospora fusca Xylanase Gene and ItsExpression in Escherichia coli and Streptomyces lividans

GURDEV S. GHANGAS, YOU-JI HU,t AND DAVID B. WILSON*

Division of Biological Sciences, Section of Biochemistry, Molecular and Cell Biology, Cornell University,Ithaca, New York 14853

Received 1 December 1988/Accepted 3 March 1989

Thermomonospora fusca chromosomal DNA was partially digested with EcoRI to obtain 4- to 14-kilobasefragments, which were used to construct a library of recombinant phage by ligation with EcoRI arms ofXgtWES.XB. A recombinant phage coding for xylanase activity which contained a 14-kilobase insert wasidentified. The xylanase gene was localized to a 2.1-kilobase Sail fragment of the EcoRI insert by subcloningonto pBR322 and derivatives of pBR322 that can also replicate in Streptomyces lividans. The xylanase activityproduced by S. lividans transformants was 10- to 20-fold higher than that produced by Escherichia colitransformants but only one-fourth the level produced by induced T. fusca. A 30-kilodalton peptide with activityagainst both Remazol brilliant blue xylan and xylan was produced in S. lividans transformants that carried the2.1-kilobase Sail fragment of T. fusca DNA and was not produced by control transformants. T. fusca cultureswere found to contain a xylanase of a similar size that was induced by growth on xylan or Solka Floc. Antiserumdirected against supernatant proteins isolated from a Solka Floc-grown T. fusca culture inhibited the xylanaseactivity of S. lividans transformants. The cloned T. fusca xylanase gene was expressed at about the same levelin S. lividans grown in minimal medium containing either glucose, cellobiose, or xylan. The xylanase bound toand hydrolyzed insoluble xylan. The cloned xylanase appeared to be the same as the major protein inxylan-induced T. fusca culture supernatants, which also contained at least three additional minor proteins withxylanase activity and having apparent molecular masses of 43, 23, and 20 kilodaltons.

Next to cellulose, hemicelluloses are the most abundantnaturally occuring polymers, and they are present in associ-ation with lignin and cellulose in plant cell walls. A majorcomponent of hemicelluloses from monocots is xylan, apolymer which consists of a backbone of 3-(1,4)-D-xylopyr-anoside residues. Xylans comprise only a minor fraction ofdicot-hemicellulose. Xylans of different origins vary withrespect to molecular weight, composition, number of sidechains, and level of acetylation.Many bacteria and fungi can utilize xylans as carbon

sources; they do so by producing xylanolytic enzymes,which are either excreted into the medium or remain asso-ciated with the cell envelope. A number of activities such asendoxylanases, exoxylanases, ,B-xylosidases, a-glucuroni-dases, a-arabinofuranosidases, and esterases have beenidentified in xylanolytic microorganisms (6, 10, 36). Out ofall the xylanolytic proteins the endoxylanases and P-xylosi-dases have been most widely studied, and xylanase genesfrom a wide variety of mesophilic bacteria and fungi havebeen cloned (4, 5, 15-17, 20, 27, 28, 31, 32, 37, 38). Studiesof a xylanase gene from an extreme thermophile werepresented in a preliminary report (P. A. Caughey, D. R.Love, and P. L. Berguist, FEMS Symp. 1987, abstr. no.P4-07, p. 90), and recently a report on the cloning of athermophilic xylanase appeared (13).Thermomonosporafusca is a filamentous soil thermophile

that produces cellulolytic, xylanolytic, and pectinolytic en-zymes (1, 14, 33, 35; D. L. Ristroph, Ph.D. dissertation,University of Pennsylvania, 1981). The xylanases of T. fusca

* Corresponding author.t Present address: Fox Chase Cancer Center, Institute for Cancer

Research, Philadelphia, PA 19111.

have been studied by several workers (1, 26, 29; Ristroph,Ph.D. dissertation).

In this report we describe the isolation by cloning intobacteriophage K of a xylanase gene from T. fusca YX. Thexylanase gene was expressed in both Escherichia coli andStreptomyces lividans. The S. lividans transformants ex-creted into the medium a 30-kilodalton (kDa) xylanase thatbound to xylan and appeared to be identical to the majorxylanase produced by xylan-induced cultures of T. fusca. Inaddition, we present initial studies on the regulation ofxylanase production in both T. fusca and S. lividans.

MATERIALS AND METHODSBacterial strains and vectors. The bacterial strains and

vectors used in this study are listed in Table 1. E. coli LE392was obtained from the E. coli Genetic Stock Center, Depart-ment of Biology, Yale University, and was used as a host forXgtWES.XB and its derivatives (21). XgtWES.XB DNA andXgtWES.XB EcoRI arms were purchased from either Be-thesda Research Laboratories, Inc., or Amersham Corp.XgtWES.XB SalI arms were prepared as described below. S.lividans TK24 and plasmid pIJ702 were kindly provided byD. A. Hopwood, John Innes Institute, Norwich, England.Growth of organisms. Unless indicated otherwise, T. fusca

YX was grown as described previously (12). E. coli strainswere grown in Luria Bertani (LB) medium (24). When usedas a host for a phage growth, LE392 was grown in 1%tryptone-2.5% sodium chloride-10 mM MgCl2-0.2% malt-ose. E. coli strains carrying recombinant plasmids weregrown in LB medium containing ampicillin (50 jig/ml).The media for S. lividans protoplast preparation and

regeneration were described previously (11, 12). Unlessindicated otherwise, S. lividans cultures were grown ontryptone soya broth as previously described (12).

2963

JOURNAL OF BACTERIOLOGY, June 1989, p. 2963-29690021-9193/89/062963-07$02.00/0Copyright © 1989, American Society for Microbiology

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

2964 GHANGAS ET AL.

TABLE 1. Bacterial strains and vectors used in this study

Bacterial strain Genotype or Referenceor vector phenotype or source

StrainsE. coliLE392 F- hsdR5J4 (rK- MK+) 24

supE44 supF58 lacYlgalK2 galT22 metBitrpR55 X-

HB101 F- hsdS20 (rK- m-) 9, 24supE44 recAJ3 ara-14lacYl proA2 galK2rpsL20 xyl-5 mtl-l X-

S. lividans TK24 str-6 D. A. HopwoodT. fusca YX36 Prototroph W. D. Bellamy

(3, 12)

VectorsXgtWES.XB 21Xxyl-1 This studyXxyl-2 This studypBR322 Ampr Tetr 8pTX101 Ampr Tets This studypIJ702 Tsrr Mel' 18pIJ702-pBR322 Tsrr Mel- 12pGG82 Tsrf Mel- This studypGG92 Tsrf Ampr This studypGG93 Tsrr Ampr This study

Recombinant DNA techniques. T. fusca chromosomalDNA and XDNA were isolated as described previously (12,24). Preparation of plasmid DNA was performed as de-scribed by Maniatis et al. (24) or by the procedure of Keiser(19).

All restriction endonucleases were purchased from Be-thesda Research Laboratories, Amersham, or New EnglandBioLabs, Inc., and restriction enzyme digests were per-formed as recommended by the manufacturer. DNA ligationwas carried out as described by Maniatis et al. (24). DNAfragments were separated on 0.7% agarose gels in 0.089 MTris-0.089 M boric acid-2 mM EDTA (pH 8.0). Aftervisualization with ethidium bromide and UV light, a strip ofDEAE membrane (S & S NA-45; Schleicher and SchuellCo.) was placed in an incision just ahead of the band.Another strip was placed just behind the band. Electropho-resis was continued until transfer was complete. The DNAon the membrane strip was eluted with 1.0 M NaCl-0.1 mMEDTA-20 mM Tris (pH 8.0). The eluted solution wasextracted with 3 volumes of water-saturated N-butanol toremove the residual ethidium bromide, and the DNA frag-ments were precipitated with 2.5 volumes of cold ethanol.

T. fusca chromosomal DNA was partially digested withEcoRI, and fragments in the range of 4 to 14 kb were isolatedby sucrose gradient centrifugation. These were ligated withXgtWES.AB EcoRI arms (purchased from Bethesda Re-search Laboratories). The ligation mixture was incubated at4°C overnight and packaged with a A in vitro packaging kit asdescribed by the manufacturer (Amersham).XgtWES.AB SalI arms were prepared by completely di-

gesting 50 jg of AgtWESAB DNA with Sall. The digestswere separated on a linear gradient of 10 to 40% sucrose(containing 1 M NaCl, 5 mM EDTA, 20 mM Tris [pH 8.0]) at27,000 rpm in a Beckman SW27 rotor at 25°C for 36 h. Thefractions near 27.6 kilobases (kb) (left arm) and 12.3 kb (rightarm) were collected, dialyzed against 10 mM Tris-1 mM

0ooLU Lu Axyl-1

10 20 30 40

c c c A xyl-210 20 30 40

EcoRI

PstI ~~~~~~SalBamHl

pTX1O1 Pst IV 6.6 Kb

Nde li\ A BamHl

AvaSall

FIG. 1. Restriction maps of XgtWES.AB and its subclones. Thicklines indicate T. fusca sequences.

EDTA (pH 8.0), and precipitated with 2.5 volumes of coldethanol. After dephosphorylation with calf alkaline phos-phatase (Boehringer Mannheim Biochemicals) by the proce-dure of Maniatis et al. (24), the DNA was ligated with apartial Sall digest of the DNA to be subcloned. The ligatedDNA was packaged with a K packaging kit.For subcloning the xylanase gene, the DNA from Xxyl-1

phage was partially digested with Sall and extracted withphenol-chloroform; the fragments were ligated withXgtWES.AB SalI arms and packaged into A phage. The insertfrom Axyl-2 was subcloned into pBR322 DNA that had beencut with Sall and dephosphorylated with calf intestine alka-line phosphatase. Transformation of E. coli was performedas described by Maniatis et al. (24).

Preparation of DNA from S. lividans, construction ofrecombinant plasmids, and protoplast transformation tech-niques were carried out as previously described (11).

Preparation of soluble xylan. Oat-spelt xylan (arabinoglu-curonoxylan) was purchased from Sigma Chemical Co.Soluble xylan (25) was prepared as follows: 10 g of xylan wasdissolved in 200 ml of H20, the pH was adjusted to 10 with1 M NaOH, the mixture was incubated at room temperaturefor 1 h and centrifuged for 10 min at 10,000 x g, and thesupernatant was neutralized with 1.0 M acetic acid andlyophilized. About 3 to 4 g of dried xylan was obtained.

Preparation of insoluble xylan. After the soluble xylan wasremoved as described above, the insoluble xylan was sus-pended in 200 ml of H20, the pH was adjusted to 7.0 with 1M acetic acid, and the preparation was filtered on Whatmanno. 1 filter paper and extensively washed with water. Thewhite paste was carefully removed and dried by lyophiliza-tion.

Screening of plaques for xylanase activity with Congo red.Plates to be screened for xylanase activity were overlaidwith 7 ml of xylan-agarose (1% agarose, 0.05% solublexylan, 10 mM Tris [pH 6.5]) and incubated at 55°C for 6 h.After incubation, the plates were stained with 0.1% Congored for 15 min. The dye solution was removed, and the plateswere destained with 1 M NaCl. A yellow ring resulting fromxylan hydrolysis appeared around positive plaques (27, 34).

Screening of colonies with RBB-xylan. Remazol brilliantblue (RBB)-xylan was prepared by the procedure of Biely et

J. BACTERIOL.

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

CLONING AND EXPRESSION OF T. FUSCA XYLANASE GENE

A pTX1O1

1. Hind lIII Sph I digest2.1solate large fragment

1. Hi2.Is

Mix ligate and transform

pGG93Pst l\

11.73 kbp

Sal I 'ShI KpnI Sal

FIG. 2. Construction of shuttle plasmids pGG92 (A) and pGG93 (B) that are capable of replicating in E. coli and S. lividans and carry T.fusca DNA coding for xylanase. Symbols: ( ) pBR322 sequence, (i) sequences originating from pIJ702, (_) Thermomonosporasequences carrying the xylanase gene. Amp, Ampicillin resistance gene expressed in E. coli; Tsr, thiostrepton resistance gene expressed inS. lividans. Only the sites that were important for this study are shown. Additional sites are shown in Fig. 1 and reference 8.

al. (7). Each plate was overlaid with 5 ml of 1% agarosecontaining 10 mM Tris (pH 6.5) and 0.07% RBB xylan andincubated at 55°C for over 4 h. If necessary, the plates weredestained with a solution of 50 mM sodium acetate (pH 5.4)in 95% ethanol (1:2, vol/vol); positive plaques were sur-rounded by a clear ring.

TABLE 2. Expression of a T. fusca xylanase gene inE. coli and S. lividans

Xylanase (U/ml)Strain

Cell extracts Culture supernatant

HB101(pTX101) 0.52 0.27HB101(pBR322) <0.01 <0.01HB101(pGG92) <0.01 <0.01HB101(pGG93) 0.22 0.15TK24(pGG82) <0.10 <0.10TK24(pGG92) 1.0 + 0.5 7.0 + 2.0

Enzyme assays. Xylanase was assayed by E. coli bygrowing a 100-ml overnight culture and harvesting it bycentrifugation (10 min at 10,000 x g); the pellet was sus-

pended in 3 ml of 0.05 M potassium phosphate buffer (pH6.5) (buffer I), and an extract was prepared in a French pressat 10,000 lb/in2. The culture supernatant was concentrated to3 ml with an Amicon UM10 membrane. Xylanase activitywas determined by measuring the rate of release of reducingsugar from xylan. Cell extract or concentrated culture su-

pernatant was mixed with buffer I to give a final volume of0.3 ml, 0.10 ml of 1% soluble xylan solution made up inbuffer I was added, and the mixture was incubated at 55°Cfor 30 min. After incubation, 0.75 ml of dinitrosalicylic acidreagent (35) was added, and the samples were incubated in a

boiling water bath for 15 min. The samples were allowed tocool to room temperature, and then the A600 was deter-mined. One unit of xylanase is defined as the amount ofenzyme that produces 1 ,umol of reducing sugar per min inthe above procedure.

B pGG82

E. coli

12.8 kbp

Saol

a

VOL. 171, 1989 2965

II

nlI

ti

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

2966 GHANGAS ET AL.

TABLE 3. Xylanase and carboxymethylcellulase activity of T.fusca grown on glucose, xylan, and cellulosea

Enzyme activity (U/ml)Carbon source

CMCase Xylanase

Glucose 0.1 1.1Oat spelt xylan 1.6 21.6Solka floc 22.8 22.4

" A 2-day culture of T. fusca in LB medium was centrifuged and suspendedin the same amount of LB medium. Then 2.0 ml of this suspension wastransferred to a 250-ml flask containing 40 ml of Hagerdal medium (22)containing 0.5% (wt/vol) of the indicated carbon source and incubated in ashaking water bath for 2 days. The supematants recovered by centrifugationwas assayed for carboxymethylcellulase (CMCase) and xylanase activities.

FIG. 3. E. coli HB101 transformed by pBR322,and pGG93. Single colonies were picked onto LBallowed to grow for about 10 h, overlaid with RBEincubated at 55°C for at least 8 h. A clear haloindicates xylanase activity. Colonies: 1, pBR322pGG92; 4, pGG93.

T. fusca and S. lividans cultures were as;

nase activity after removing the mycelia eitlgation or by filtration through glass wool. (cellulase activity was determined by replaccarboxymethylcellulose in the above assay (

B 1 2A

-25.7

-18

-14

-6

FIG. 4. Comparison of the cloned xylanase and the T. fuscaxylanase. (A) Activity-stained SDS gel. Supernatants from T. fusca(4 ,ul) and S. lividans (40 ,Jl) were mixed with 10 1.l of loading dyeand boiled for 5 min. The boiled samples (30 p.l) were electro-phoresed on an SDS-12.5% polyacrylamide gel. SDS was removedfrom the gel as described previously (2) and incubated with a xylanagar overlay for 60 min in a 55°C oven. The overlay was stained withCongo red, the acrylamide gel was then stained with Coomassie blueto locate the protein standards, and the two gels were alligned tocalculate the molecular weights of the activity bands. Lanes; 1, T.fusca (xylan grown); 2, mixture of T. fusca and S. lividans(pGG92);3, S. Iividans(pGG92); 4, 5. lividans(pGG82); 5, 5. lividans(pIJ702).(B) Coomassie blue-stained SDS gel. Lanes: 1, T. fusca xylan-grownculture supernatant (0.5 U); 2, mixture of T. fusca xylanase (0.25 U)and S. lividans xylanase concentrated by 50% (NH4)2SO4 precipita-tion (0.25 U); 3, S. lividans xylanase concentrated by 50%(NH4)2SO4 precipitation (0.5 U).

Xylanase binding assays. Insoluble xylan (from 0 to 20 mg)r 4 was added to 1.5-ml microcentrifuge tubes and washed two

ol times with 1.0 ml of water by centrifugation for 30 s andplaced on ice. Then 0.5 ml of transformed S. lividans culture

pTX.101, pGG92, supernatant was added to each tube, tubes were incubatedampicllin plates, for 5 min and centrifuged, and the supernatant was assayedB-xylan agar, and for xylanase. The activity lost from the supernatant wasaround a colony assumed be the activity bound.

2,pTX1'1; ' Western blotting. Western blotting (immunoblotting) wascarried out as previously described (12). Briefly, proteinswere separated on sodium dodecyl sulfate (SDS) gels (12.5%

sayed for xyla- polyacrylamide), transfered to nitrocellulose, and reactedher by centrifu- with the appropriate rabbit antiserum; primary antigen-Carboxymethyl antibody complexes were detected by reacting with an

Sing xylan with anti-rabbit immunoglobulin G-coupled alkaline phosphatase35). and visualized with nitroblue tetrazolium chloride and 5-

bromo-4-chloro-3-indolylphosphate.Other procedures. Electrophoresis of proteins on poly-

3 4 acrylamide gels or SDS-polyacrylamide gels were performedas described before (11). The zymogram technique for Be-guin developed for cellulase was adapted to locate xylanasebands in SDS gels (2, 23). The bands were located by Congored staining of xylan-agar and/or clearing of RBB-xylan agar

-43 replicas. Antibody inhibition assays were carried out as

previously described (11).

RESULTS

Xylanase gene cloning. The transformation efficiency of therecombinant Agt phage T. fusca genome bank was 8 x 106

PFU/,ug of DNA. Of 3,500 plaques screened for xylanaseactivity by both the Congo red and the RBB-xylan proce-dures, 2 showed xylanase activity. Based on restrictionmaps, the DNA of both phage contained the same -14-kbEcoRI insert. This recombinant phage was designated Axyl-1(Fig. 1).

Subcloning of the xylanase gene. Since there were no

EcoRI sites in the insert DNA, Xxyl-1 DNA was subclonedby using XgtWES.AB Sall arms as described in Materials andMethods. About 1,500 plaques were screened by the Congored procedure, and 11% were xylanase positive. Sevenpositive plaques were purified, and their xylanase activitiesand restriction maps were determined. One phage had aninsert DNA containing 3.4- and 2.1-kb SalI fragments,whereas all of the other phage contained both of thesefragments along with at least one additional fragment. Thexylanase-positive phase with the smallest insert was calledXxyl-2, and its restriction map is shown in Fig. 1.Xxyl-2 DNA was digested with SalI, and the 3.4- and

2.1-kb fragments were separated by gel electrophoresis andsubcloned into pBR322. Transformants were screened for

>: 3

-4

e

J. BACTERIOL.

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

CLONING AND EXPRESSION OF T. FUSCA XYLANASE GENE

2 3 4

-43

-25.7

-18

~

u S 6

-3

FIG. 5. SDS-gel electrophoretic analysis of xylanase induction inT. fusca. Cells grown in LB medium were used to innoculateHagerdal medium containing the indicated carbon sources (0.5%) ata dilution of 1:20 and incubated in a shaker at 55°C. After 24 h thesupernatants were prepared by centrifugation (10,000 x g for 10min), assayed, electrophoresed on an SDS-12.5% polyacrylamidegel, and stained with Coomassie brilliant blue. Each lane contains 16,ul of supernatant. Lanes: 1, glucose grown (0.4 U of xylanase per

ml); 2, xylan grown (15 U of xylanase per ml); 3, cellulose grown (13U of xylanase per ml).

xylanase activity by the Congo red procedure. All of thecolonies from the 3.4-kb fragment ligation were negative,whereas 35% of the transformants from the 2.1-kb ligationwere positive. Restriction maps of several transformantsshowed that they all contained only the 2.1-kb insert. Thisplasmid was designated pTX101, and its restriction map isalso shown in Fig. 1.

Construction of E. coli-S. lividans shuttle plasmids contain-ing the xylanase gene. The multicopy plasmid pIJ702 (18)commonly used for cloning in S. lividans contains five Sallrestriction sites and therefore is not convenient for cloningSall fragments. Instead, a derivative of pIJ702, pGG82., wasconstructed and used for expressing the T. fusca xylanasegene in S. lividans. In plasmid pGG82 the smaller BglII-SphIfragment of pIJ702 is replaced by a BamHI-SphI fragment ofE. coli plasmid DNA. The structure of pGG82 appears in

Fig. 2. The EcoRI-SphI fragment of pGG82 is from pBR322,and the BamHI-EcoRI fragment is derived from pUR222(30). Plasmid pGG82, unlike plasmid pIJ702, no longerproduces melanin (Mel- phenotype) and contains singleEcoRI and HindIll sites. The BglII site of pIJ702 is inacti-vated in pGG82, due to its fusion to the BamH-I site ofpUR222.

Further restriction analysis of pTX101 showed that it hadno sites for SstI and BglII, one site each for EcoRI, SphI,Hindlll, and BalI, and three sites for Scal. The constructionof shuttle plasmid pGG92 from pTX101 and pGG82 isdepicted in Fig. 2A. The three E. coli HB101 transformantstested had the correct restriction pattern. This plasmid isstable in E. coli HB101. The construction of shuttle plasmidpGG93 from pGG82 and pTX101 is shown in Fig. 2B. The

TABLE 4. Xylanase production in S. lividans cultures grownin glucose and xylan'

Xylanase in culturePlasmid Carbon source filtrate (U/ml) on day:contained

2 4

pGG82 Glucose 0.1 0.1Xylan 0.0 2.6

pGG92 Glucose 1.1 2.8Xylan 0.0 5.0

"Transformed colonies were grown in tryptone soya broth containingthiostrepton. Two-day cultures (mycelial volume, about 10%0) were centri-fuged and taken up in modified Hagerdal medium (22) lacking a carbon source.Then 1 ml of this culture was transferred to 50 ml of the indicated mediumcontaining 10 p.g of thiostrepton per ml and 0.25 g (0.5%) of either glucose orxylan. The cultures were shaken at 30°C, and at the indicated times sampleswere removed, centrifuged, and assayed for xylanase activity.

12.8-kb plasmid DNA was isolated from six E. coli transfor-mants, and all six had the orientation shown in Fig. 2.

Expression of shuttle plasmids in E. coli. Unlike plasmidpTX101, which had strong xylanase activity in E. coli HB101(Table 2), shuttle plasmid pGG92 failed to confer a xylanase-positive phenotype on HB101 (Fig. 3). The tetracyclineresistance promoter is deleted during the construction ofpGG92 while leaving the T. fusca DNA sequence intact, andwhen this promoter was present as in pGG93, xylanaseactivity was present in the HB101 transformants.

Transformation of S. lividans and expression of the T. fuscaxylanase gene. Plasmid DNA was isolated from E. colitransformed with pGG92 and pGG93 and was used totransform S. lividans protoplasts. pGG92 consistently gavemore transformants than pGG93 when equal amounts ofDNA were used. Furthermore, all of the pGG92 transfor-mants were xylanase positive, whereas only a few of thepGG93 transformants were xylanase positive, possibly be-cause most pGG93 transformants were unstable due torecombination between the duplicated regions of pBR322present in pGG93. The xylanase activities of transformedand control strains are given in Table 2.When supernatant proteins from S. lividans transformants

were separated on an SDS gel and stained for xylanaseactivity, a 30-kDa protein was the major active band. Thisband was not present in control cultpres (Fig. 4).The xylanase in S. lividans transformants is similar to an

induced xylanase in T. fusca. Table 3 shows the specificinduction of extracellular xylanase activity in T. fusca cul-tures grown for 2 days on xylan. Xylan-grown cultures hadlow levels of cellulase and high levels of xylanase activity,whereas cellulose (Solka floc)-grown cultures had high levelsof both activities. The level of xylanase was higher than thelevel of cellulase in glucose-grown cultures. The proteinspresent in the supernatants of cultures grown on the differentcarbon sources were separated by gel electrophoresis (Fig.5). The major protein present in the supernatant of xylan-grown cells migrated in the same position as the clonedxylanase (Fig. 4). Four bands of xylanase activity werepresent in stained gels (43, 30, 23, and 20 kDa) run on the T.fusca sample. The larger two (43 and 30 kDa) are probablydistinct gene products, whereas the smaller forms may bederived by proteolysis of the larger enzymes, especially asthe lower-molecular weight bands are similar in size to theproteolysis products present in S. lividans transformants.The xylanase activity of S. lividans transformants was

inhibited by an antiserum prepared against proteins isolated

VOL. 171, 1989 2967

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

2968 GHANGAS ET AL.

0.Al 2 3 4 5 B5 4 3 2 1 -

,

0.

N

00(0(a0)

N-

U0._

0)U) O.,"*0

cn

.rio0.0

E:. 0.,

FIG. 6. Comparison of cloned xylanase and T. fusca proteins byactivity staining (A) and Western blotting (B). Supernatants from T.fusca and S. lividans transformant cultures were diluted withloading dye and left in a boiling water bath for 5 min. The boiledsamples were subjected to electrophoresis on an SDS-12.5% poly-acrylamide gel to separate the peptides. After SDS was removed as

described in the text, the gel was incubated with xylan agar replicafor 30 min in a 55°C oven. The replica was stained with Congo red.Proteins in the acrylamide gel were electroblotted onto nitrocellu-lose (0.45-,um pore size) and probed with antiserum directed againstSolka Floc-grown culture supematant proteins. Lanes: 1, T. fuscaculture grown in glucose; 2, T. fusca culture grown in xylan; 3, T.fusca culture grown in Solka floc; 4, S. Iividans(pGG82) transform-ant culture; 5, S. lividans(pGG92) transformant culture. Std, proteinmolecular mass standards of 43, 25.7, 18, and 14 kDa.

from the culture supernatant of cellulose-grown T. fusca.The activity was not inhibited by control serum or serum

prepared against purified cellulases from T. fusca (data notshown). Immunoblotting of S. lividans transformants re-vealed a 30-kDa protein; control cultures lacked this band(Fig. 6).

Regulation of xylanase production of S. lividans transfor-mants and its binding to xylan. S. lividans transformantscontaining the control plasmid (pGG82) did not produce hostxylanase in medium lacking xylan (Table 4), whereas theplasmid-encoded T. fusca xylanase was produced in minimalmedium containing glucose, cellobiose, or xylan and intryptone soya broth. The activity of pGG92 cultures grownon xylan was about two- to threefold the activity of controlcultures and contained both induced host activity and T.fusca xylanase.

The lack of xylanase activity at early time points in thesupernatants of cultures grown on xylan could indicate thatthe activity binds to undegraded insoluble xylan. In fact,xylanase activity quickly disappeared from the solution afterthe addition of insoluble xylan, and the bound enzymereleased soluble reducing sugars at 55°C in 50 mM KP, (pH6.5) (Fig. 7).

DISCUSSIONThe cloned T. fusca xylanase gene was expressed at a low

level in E. coli and at a higher level in S. lividans, which also

1 2 3time of incubation (hrs)

4

FIG. 7. Release of soluble reducing sugars into the medium bycloned xylanase bound to insoluble xylan. Supernatant (0.5 ml) froman S. Iividans(pGG92) transformant culture (5 U.of xylanase per ml)grown in tryptone soya broth was added to a 1.5-mil plastic tube on

ice containing 20 mg of insoluble xylan. The sample was gentlymixed for 5 min at 4°C and centrifuged to separate the xylan andsupernatant. The xylanase activity in the supernatant was 0.6 U/ml(88% binding). The xylan was washed twice with 1 ml of cold H20,mixed with 0.5 ml of 0.05 M potassium phosphate buffer (pH 6.5),and incubated at 55°C. A control S. lividans(pGG82) culture was

treated similarly. At appropriate times the tubes were centrifuged toremove 10-1udl samples for the determination of reducing sugar.Symbols: A, pGG92 transformant; 0, pGG82 transformant.

efficiently excretes the activity into the culture medium. Theresults of expression in S. lividans indicate that the xylanasegene uses its own promoter. However, efficient expressionin E. coli seems to require additional promoter activity thatmay be provided by the adjoining T. fusca or X DNA on theXxyl-1 and Xxyl-2 phages or the tet promoter in pTX101 orpGG93. The xylanase produced by the T. fusca gene presentin plasmid pGG92 clearly differed from S. lividans xylanaseactivity in its regulation, electrophoretic mobility, and reac-tion with T. fusca antiserum.

S. lividans xylanase activity is induced by xylan; an S.

lividans xylanase gene cloned into S. lividans on plasmidpIJ702 was reported to be inducible (27). However, the T.fusca xylanase gene described here was not induced byxylan in S. lividans, although it was induced in T. fusca.A number of xylanase genes from different organisms have

been expressed in E. coli; however, this report appears to bethe first example of heterologous xylanase expression in S.lividans. These studies demonstrate the usefulness of the S.lividans system for expressing T. fusca genes (11, 12).Streptomyces transformants that carry the T. fusca xyla-

nase gene produce a 30-kDa protein that displays xylanaseactivity and appears in all respects to be the same as themajor protein in xylan-induced T. fusca supernatants. Thexylanase activity of Streptomyces transformants can bespecifically and quantitatively inhibited by an antiserumraised against induced T. fusca extracellular protein. Theavailability of large amounts of xylanase protein in S.lividans culture supernatants free from other T. fusca pro-teins and the xylan-xylanase binding demonstrated in this

.5.s X~~~~~~pGG92

4.

,3

.2-

pGG82

O _&I W0

J. BACTERIOL.

0

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

CLONING AND EXPRESSION OF T. FUSCA XYLANASE GENE

study should facilitate purification and structure-functionstudies of this protein.

ACKNOWLEDGMENTS

This work was supported by U.S. Department of Energy grantFG02-84ER13233 and by grant 85-CRCR-1-1880 from the U.S.Department of Agriculture.We thank S. J. Lucania, Squibbs Institute, Princeton, N.J., for

the thiostrepton.

LITERATURE CITED1. Ball, A. S., and A. J. McCarthy. 1988. Saccharification of straw

by Actinomyces enzymes. J. Gen. Microbiol. 134:2139-2147.2. Beguin, P. 1983. Detection of cellulase activity in polyacryl-

amide gels using Congo Red-stained agar replicas. Anal. Bio-chem. 131:333-336.

3. Bellamy, W. D. 1977. Cellulose and lignocellulose digestion bythermophilic acinomycetes for single cell protein production.Dev. Ind. Microbiol. 18:249-254.

4. Bernier, R., Jr., and M. Desrochers. 1985. Molecular cloning ofa beta xylosidase gene from Bacillus subtilis. J. Gen. Appl.Microbiol. 31:513-518.

5. Bernier, R., Jr., H. Driguez, and M. Desrochers. 1983. Molecu-lar cloning of a Bacillus subtilis xylanase EC.3.2.1.8 gene inEscherichia coli. Gene 26:59-66.

6. Biely, P. 1985. Microbial xylanolytic systems. Trends Biotech-nol. 3:286-290.

7. Biely, P., D. Mislovicova, and R. Toman. 1985. Soluble chro-mogenic substrates for the assay of endo-1,-4-,-xylanases andendo-1,4-0-glucanases. Anal. Biochem. 144:142-146.

8. Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L.Heyneker, and H. B. Boyer. 1977. Construction and characteri-zation of new cloning vehicles. II. A multipurpose cloningsystem. Gene 2:95-113.

9. Boyer, H. B., and D. Rouland-Dussoix. 1969. A complementa-tion analysis of the restriction and modification of DNA inEscherichia coli.J. Mol. Biol. 41:459-472.

10. Dekker, R. F. H. 1985. Biodegradation of the hemicelluloses, p.505-533. In T. Higuchi (ed.), Biosynthesis and biodegradationof wood components, Academic Press, Inc., New York.

11. Ghangas, G. S., and D. B. Wilson. 1987. Expression of aThermomonospora fusca cellulase gene in Streptomyces liv-idans and Bacillus subtilis. Appl. Environ. Microbiol. 53:1470-1475.

12. Ghangas, G. S., and D. B. Wilson. 1988. Cloning of the Ther-momonospora fusca endoglucanase E2 gene in Streptomyceslividans: affinity purification and functional domains of thecloned gene product. Appl. Environ. Microbiol. 54:2521-2526.

13. Grepinet, 0, M.-C. Chebrou, and P. Beguin. 1988. Purificationof Clostridium thermocellum xylanase Z expressed in Esche-richia coli and identification of the corresponding product in theculture medium of C. thermocellum. J. Bacteriol. 170:4576-4581.

14. Hagerdal, B. G. R., J. D. Ferchak, and E. K. Pye. 1978.Cellulolytic enzyme system of Thermoactinomyces sp. grownon microcyrstalline cellulose. Appl. Environ. Microbiol. 36:606-612.

15. Honda, H., T. Kudo, and K. Horikoshi. 1985. Selective excre-tion of alkaline xylanase by Escherichia coli carrying P-CX-311.Agric. Biol. Chem. 49:3011-3016.

16. Honda, H., T. Kudo, and K. Horikoshi. 1986. Production ofextracellular alkaline xylanase of alkalophilic bacillus-SP C-125by Escherichia coli carrying PCX-311. Syst. Appl. Microbiol.8:152-157.

17. Iwasaki, A., H. Kishida, and M. Okanishi. 1986. Molecularcloning of a xylanase gene from Streptomyces-SP No. 36A andits expression in Streptomycetes. J. Antibiot. 39:985-993.

18. Katz, E., C. J. Thompson, and D. A. Hopwood. 1983. Cloningand expression of the tyrosinane gene from Streptomyces anti-bioticus in Streptomyces lividans. J. Gen. Microbiol. 129:2703-2714.

19. Kieser, T. 1984. Factors affecting the isolation of CCC DNAfrom Streptomyces lividans and Escherichia coli. Plasmid 12:19-36.

20. Kudo, T., A. Ohkoshi, and K. Horikoshi. 1985. Expression of axylanase gene of alkalophilic aeromonas-SP. No. 212 in Esche-richia coli. J. Gen. Microbiol. 131:2825-2830.

21. Leder, P., D. Tiemeier, and L. Enquist. 1977. EK2 derivatives ofbacteriophage lambda useful in the cloning of DNA from higherorganisms: the Agt WES system. Science 196:175-178.

22. Lin, E., and D. B. Wilson. 1987. Regulation of ,B-1,4-endoglu-canase synthesis in Thermomonospora fusca. Appl. Environ.Microbiol. 53:1352-1357.

23. MacKenzie, C. R., and R. E. Williams. 1984. Detection ofcellulase and xylanase activity in isoelectric-focused gels usingagar substrate gels supported on plastic film. Can. J. Microbiol.30:1522-1525.

24. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, New York.

25. Matsuo, M., T. Yasui, and T. Kobayashi. 1977. Purification andsome properties of 3-xylosidase from Malbranchea pulchellavar. sulfurea no. 48. Agric. Biol. Chem. 41:1593-1599.

26. McCarthy, A. J., E. Peace, and P. Broda. 1985. Studies on theextracellular xylanase activity of some thermophilic actino-mycetes. Appl. Microbiol. Biotechnol. 21:238-244.

27. Mondou, F., F. Sharech, R. Morosoli, and D. Kluepfel. 1986.Cloning of the xylanase gene of Streptomyces lividans. Gene49:323-330.

28. Panbangred, W., T. Kondo, S. Negoro, A. Shinmyo, and H.Okada. 1983. Molecular cloning of the genes for xylan degrada-tion of Bacillus pumilus and their expression in Escherichia coli.Mol. Gen. Genet. 192:335-341.

29. Ristroph, D. L., and A. E. Humphrey. 1985. Kinetic character-ization of the extracellular xylanases of Thermomonospora sp.Biotechnol. Bioeng. 27:832-836.

30. Ruther, U., M. Koenen, K. Otto, and B. Muller-Hill. 1981.pUR222, a vector for cloning and rapid chemical sequencing ofDNA. Nucleic Acids Res. 9:4087-4098.

31. Sandhu, J. S., and J. F. Kennedy. 1984. Molecular cloning ofBacillus polymyxa 1-4-beta-D xylanase EC-3.2.1.8 gene in Esch-erichia coli. Enzyme Microb. Technol. 6:271-274.

32. Sipat, A., K. A. Taylor, R. Y. C. Lo, C. W. Forsberg, and P. J.Krell. 1987. Molecular cloning of a xylanase gene from Bacteroi-des succinogens and its expression in Escherichia coli. Appl.Environ. Microbiol. 53:477-481.

33. Stutzenberger, F. J. 1987. Inducible thermoalkalophilic polyga-lacturonate lyase from Thermomonospora fusca. J. Bacteriol.169:2774-2780.

34. Teather, R. M., and P. J. Wood. 1982. Use of Congo red-polysaccharide interactions in enumeration and characterizationof cellulolytic bacteria from bovine rumen. Appl. Environ.Microbiol. 43:777.

35. Wilson, D. B. 1988. Cellulases of Thermomonospora fusca.Methods Enzymol. 160:314-323.

36. Woodward, J. 1984. Xylanases: functions, properties and appli-cations. Top. Enzyme Ferment. Biotechnol. 8:9-30.

37. Yang, R. C. A., C. R. MacKenzie, D. Bilous, V. L. Seligy, andS. A. Narang. 1988. Molecular cloning and expression of axylanase gene from Bacillus polymyxa in Escherichia coli. Appl.Environ. Microbiol. 54:1023-1029.

38. Zappe, H., D. T. Jones, and D. R. Woods. 1987. Cloning andexpression of a xylanase gene from Clostridium acetobutylicumP262 in Escherichia coli. Appl. Microbiol. Biotechnol. 27:57-63.

VOL. 171, 1989 2969

on March 16, 2018 by guest

http://jb.asm.org/

Dow

nloaded from