Molecular Cloning, Sequencing, and Expression of a Novel ...

2 downloads 0 Views 474KB Size Report
ISAAC K. O. CANN,† SVETLANA KOCHERGINSKAYA, MICHAEL R. KING, BRYAN A. WHITE,. AND RODERICK I. MACKIE*. Department of Animal Sciences, ...
JOURNAL OF BACTERIOLOGY, Mar. 1999, p. 1643–1651 0021-9193/99/$04.0010 Copyright © 1999, American Society for Microbiology. All Rights Reserved.

Vol. 181, No. 5

Molecular Cloning, Sequencing, and Expression of a Novel Multidomain Mannanase Gene from Thermoanaerobacterium polysaccharolyticum ISAAC K. O. CANN,† SVETLANA KOCHERGINSKAYA, MICHAEL R. KING, BRYAN A. WHITE, AND RODERICK I. MACKIE* Department of Animal Sciences, University of Illinois at Urbana-Champaign Urbana, IL 61801, USA Received 21 August 1998/Accepted 16 December 1998

The manA gene of Thermoanaerobacterium polysaccharolyticum was cloned in Escherichia coli. The open reading frame of manA is composed of 3,291 bases and codes for a preprotein of 1,097 amino acids with an estimated molecular mass of 119,627 Da. The start codon is preceded by a strong putative ribosome binding site (TAAGGCGGTG) and a putative 235 (TTCGC) and 210 (TAAAAT) promoter sequence. The ManA of T. polysaccharolyticum is a modular protein. Sequence comparison and biochemical analyses demonstrate the presence of an N-terminal leader peptide, and three other domains in the following order: a putative mannanase-cellulase catalytic domain, cellulose binding domains 1 (CBD1) and CBD2, and a surface-layer-like protein region (SLH-1, SLH-2, and SLH-3). The CBD domains show no sequence homology to any cellulose binding domain yet reported, hence suggesting a novel CBD. The duplicated CBDs, which lack a disulfide bridge, exhibit 69% identity, and their deletion resulted in both failure to bind to cellulose and an apparent loss of carboxymethyl cellulase and mannanase activities. At the C-terminal region of the gene are three repeats of 59, 67, and 56 amino acids which are homologous to conserved sequences found in the S-layer-associated regions within the xylanases and cellulases of thermophilic members of the Bacillus-Clostridium cluster. The ManA of T. polysaccharolyticum, besides being an extremely active enzyme, is the only mannanase gene cloned which shows this domain structure. modular organization and consist usually of a single catalytic domain linked to one or more noncatalytic domains. However, bifunctional polysaccharidases comprising two dissimilar catalytic domains have been identified via gene cloning in Ruminococcus flavefaciens (13, 44), Clostridium thermocellum (1), C. saccharolyticus (14, 28), and Anaerocellum thermophilum (45). We have recently isolated a number of thermophilic bacteria from the waste pile of a canning factory in Illinois. Among these isolates are two organisms closely related at the 16S ribosomal DNA level (98% similarity; GenBank accession numbers U40229 and U75993). These organisms, which have an optimum temperature for growth of 68°C, are highly saccharolytic and produce a number of polysaccharide-hydrolyzing enzymes. In this work, we describe the cloning, sequencing, and expression of a gene coding for a unique modular protein exhibiting both mannanase and endoglucanase activities from Thermoanaerobacterium polysaccharolyticum, one of these novel bacteria. Furthermore, we show that this large enzyme (119.6 kDa), which contains a unique catalytic region, duplicated cellulose-binding domains (CBD), and a triplicated surface-layer-like protein (SLH) domain within the same polypeptide, is encoded by a single gene.

Thermophilic anaerobic bacteria have attracted much interest for use in the bioconversion of industrial and agricultural lignocellulosic waste materials into value-added chemicals. The past two decades have witnessed a surge in exploiting microbial enzymes to unlock the potential energy in lignocellulosic biomass, such as agricultural byproducts and crop residues (31, 37). Besides the focus on polysaccharide depolymerases to hydrolyze plant cell wall polymers such as cellulose and hemicellulose and their subsequent fermentation to alcohol, there is also interest in the application of microbial enzymes to the paper industry (7, 41). Preliminary results show that the addition of xylanases and b-mannanases can reduce the loading of chlorine-containing chemicals during the bleaching of pulp (7), hence reducing the cost of chemical waste disposal. Endo-1,4-b-D-mannanases (EC 3.2.1.78) catalyze the random hydrolysis of the b-D-1,4-mannopyranosyl linkages within the main chain of mannans and various polysaccharides consisting mainly of mannose, which are components of the plant cell wall. Thermostable mannanases have been purified in Bacillus stearothermophilus (36), as well as cloned and sequenced in Caldocellulosirruptor saccharolyticus (14, 24, 28). In order to utilize plant material as a source of carbon and energy, bacteria produce an array of hydrolytic enzymes with various specificities, which act cooperatively to convert these substrates to their constituent sugars (38). These enzymes commonly show a

MATERIALS AND METHODS Bacterial strains and culture conditions. T. polysaccharolyticum was isolated from subsurface samples from the leachate of the waste pile from the canning factory in Hoopeston, Ill. The bacterium was routinely grown in a minimal medium with glucose as the energy source at 68°C. The isolation and characterization of T. polysaccharolyticum has been described elsewhere (8). Escherichia coli strains were grown at 37°C in Luria-Bertani (LB) medium supplemented with either ampicillin (100 mg/ml) or kanamycin sulfate (50 mg/ml). The reagents isopropyl-b-D-thiogalactopyranoside (IPTG) and 5-bromo-4-chloro-3-indolyl-bD-galactopyranoside (X-Gal) were added to the media where appropriate at concentrations of 125 and 80 mg/ml, respectively. DNA isolation and manipulation. Genomic DNA from T. polysaccharolyticum was isolated by the method of Marmur (25). Recombinant plasmids were ex-

* Corresponding author. Mailing address: University of Illinois at Urbana-Champaign, 458 Animal Sciences Laboratory, 1207 W. Gregory Dr., Urbana, IL 61801. Phone: (217) 244-2526. Fax: (217) 3338809. E-mail: [email protected]. † Present address: Department of Molecular Biology, Biomolecular Engineering Research Institute, 6-2-3 Furuedai, Suita, Osaka 565-0874 Japan. 1643

1644

CANN ET AL.

tracted as described by Birnboim and Doly (6). Isolation of recombinant plasmid for DNA sequence analysis was performed by using the Qiagen Midi-Prep System (Qiagen, Inc., Chatsworth, Calif.). Restriction enzymes were purchased from Stratagene (Stratagene Cloning Systems, La Jolla, Calif.) and were used according to the manufacturer’s instructions. Cloning of manA gene of T. polysaccharolyticum. Total genomic DNA was extracted from T. polysaccharolyticum cells grown to mid-exponential phase and was then partially digested with EcoRI. DNA fragments ranging from 3 to 12 kb were ligated into EcoRI-digested and dephosphorylated Lambda Zap Express arms (Stratagene). The resulting recombinant phage particles were packaged, and the products were incubated with E. coli XL-1 Blue cells {D(mcrA)183 D(mcrCB-hsdSMR-mrr)173 endA1 supE44 thi-1 recA1 gyrA96 relA1 lac [F9 proAB lacIqZDM15 Tn10(Tetr)]} at 37°C with gentle shaking. The mixture was plated on NZY medium in a soft agar containing the chromogenic substrate Ostazin brilliant red-hydroxyethyl cellulose (OBR-HEC) and incubated at 37°C overnight. The recombinant phages exhibiting endoglucanase activity produced halos within a red background. Endoglucanase-positive phages were cored and purified, and the circularized phagemid with the DNA insert coding for the ManA of T. polysaccharolyticum was recovered through an f1 helper phage-mediated in vivo excision and recircularization in E. coli XLOLR cells {D(mcrA)183, D(mcrCB-hsdSMR-mrr)173 endA1 thi-1 recA1 gyrA96 relA1 lac [F9 proAB lacIq ZDM15 Tn10(Tetr)] Su2 (nonsuppressing) lr (lambda resistant)}. DNA sequencing. The nucleotide sequences of both strands of the DNA insert (3.4 kb) were determined by the University of Illinois Biotechnology Center by using an Applied Biosystems 373A automated DNA sequencer (Applied Biosystems, Foster City, Calif.) with both dye primers and dye dideoxynucleotide chemistries. Computer analysis of DNA sequences was performed by using the DNA sequence analysis software packages Geneworks 2.45 (Intelligenetics, Inc., Mountain View, Calif.) and CLUSTAL V (16). Homology searches in the GenBank were carried out by using the BLAST program (2). Genome-walking PCR. Genome-walking PCR was used to obtain sequence downstream from of the 3.4-kb insert since sequence analysis of the 3.4-kb insert did not reveal an in-frame termination codon. The procedure for genomewalking PCR was as previously described (28). A primer targeted from bases 2896 to 2916 (59-GGATGGCCTGATTGGGGTTAT-39) of the T. polysaccharolyticum genomic DNA insert was designed to perform a genome-walking PCR to obtain a sequence from downstream of the 3.4-kb insert. An aliquot (500 ng in a 50-ml reaction) of genomic DNA from T. polysaccharolyticum was completely digested with BamHI. Two complementary oligonucleotides, the upper linker with a BamHI sticky end (59-GATCGCGCAGGAAACAGCTATGACCGGT39) and the lower linker (59-ACCGGTCATAGCTGTTTCCTGCGC-39) were synthesized. Portions (75 pmol) of each oligonucleotide in a 50-ml reaction mixture were assembled into a BamHI linker by denaturation at 94°C and annealing at 50°C for 30 min. The linker library was constructed by ligating 1 ml of the linker to 5 ml of BamHI-digested genomic DNA in a volume of 10 ml overnight at 4°C. The volume of the library was increased to 50 ml, and 5 ml was used as the template to amplify, via PCR, the region downstream of the 3.4-kb insert. The PCR mixture contained 250 mM concentrations of each deoxynucleoside triphosphate, 2 mM MgCl2, and other ingredients as described by the manufacturer (LA PCR kit; TaKaRa Shuzo Co., Ltd., Kyoto, Japan). The oligonucleotides were synthesized by standard methods with an automated DNA/ RNA synthesizer (Applied Biosystems). PCR was performed in a programmable thermal controller (PTC-100; MJ Research, Inc., Watertown, Mass.). The thermal profiles involved 29 cycles of denaturation at 94°C for 30 s, annealing at a temperature of 58°C for 50 s, and extension at 72°C for 1.5 min. The PCR fragment obtained was ligated into a pGEM-T vector DNA (Promega, Madison, Wis.), and the product was used in transforming E. coli JM109 cells through electroporation. The insert was subsequently sequenced as described above. The amplification of the complete manA gene was achieved by using the oligonucleotides 59-TGTTGAGCCGGCCTAGCGACAACG-39 (bases 86 to 103) and 59CCGCCTGAAACCTTTATGCC-39 (bases 4120 to 4101) as the forward and reverse primers, respectively. T. polysaccharolyticum genomic DNA served as the template. Deletion of CBD1 and CBD2. The two oligonucleotides 59-TGTTGAGCCG GCCTAGCGACAACG-39 (nucleotides 86 to 103) and 59-GTCTTGCCAGCT GTCCAGACCGTC-39 (nucleotides 2479 to 2455) as forward and reverse primers, respectively, were used to amplify a fragment of manA with both CBDs and the surface-layer-protein-like sequences (SLH) deleted. The PCR fragment was cloned into pGEM-T vector as described above. Four positive clones harboring the insert were used to assay for both cellulose binding and enzymatic activities. Purification of recombinant mannanase and enzyme assay. E. coli JM109 cells [endA1 recA1 gyrA96 thi hsdR17 relA1 supE44 D(lac-proAB) (F9 traD36 proAB lacIq ZDM15)] containing the manA gene were grown overnight to saturation in 1.5 liters of LB broth supplemented with ampicillin at 100 mg/ml. The cells were harvested by centrifugation at 10,000 3 g for 20 min at 4°C followed by suspension in 10 ml of 50 mM sodium phosphate buffer (pH 7.0) containing 0.1 mM phenylmethylsulfonyl fluoride. The disruption of cells to release recombinant protein was achieved by two passages through a French pressure cell at 16,000 lb/in2. Cell debris was removed by centrifugation at 10,000 3 g for 20 min at 4°C. E. coli proteins were partially removed by heat precipitation at 70°C for 10 min followed by centrifugation as described above. The supernatant from the heatprecipitated sample was applied to a column packed with Sigmacell Type 50

J. BACTERIOL. (Sigma Chemical Co., St. Louis, Mo.) which had been equilibrated with 50 mM sodium phosphate buffer (pH 5.8). To elute unbound proteins, the column was washed continuously with the equilibration buffer while monitoring the protein elution at 280 nm by spectroscopy (Beckman DU7500). The elution of bound protein was carried out with distilled water, and the proteins were concentrated (Centricon 10; Amicon, Beverly, Mass.) and resuspended in a volume of 50 mM sodium phosphate buffer (pH 7.0). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and detection of enzyme activity. The detection of activity by plate assay was done by growing E. coli cells harboring the recombinant plasmids with the manA gene on LB plates supplemented with the appropriate antibiotic. The cells were grown overnight and lysed in a chloroform chamber for 4 to 5 min. The plates were then flooded with a top agar containing 0.8% agarose and 0.5% carboxymethyl cellulose (CMC; Sigma). After a setting period, the plates were incubated upright at 65°C overnight and stained with a solution of 0.1% congo red for 2 min. This was followed by destaining through repeated washings with 1 M NaCl. Zymograms were prepared by using 0.1% CMC or galactomannan copolymerized with polyacrylamide. After separation, SDS was removed by repeated washings with ethanol-water (1:1 [vol/vol]). The gel was then equilibrated through several washings and finally immersion in 50 mM sodium acetate buffer (pH 6.0). This was followed by incubation at 65°C for 2 h, staining with 0.1% congo red, and destaining in 1 M NaCl (40). Mannanase and endoglucanase activities were determined by measuring the enzymatic release of reducing groups from locust bean gum and CMC, respectively. Standard assay mixtures (total volume, 500 ml) contained 450 ml of 0.5% substrate in 50 mM sodium phosphate buffer (pH 5.7) and 50 ml of appropriately diluted enzyme. After an incubation period of 10 min at 65°C, the reaction was terminated by the addition of 1.5 ml of ice-cold solution of 0.1% para-hydroxybenzoic acid hydrazide containing 0.4 M NaOH and 100 mM sodium citrate (20). The reaction mixture was boiled for 10 min in a water bath, and the absorbance was measured at a wavelength of 410 nm. Glucose was used as the standard in the assay, and each reaction and its control were run in triplicate. SDS-PAGE was performed with 10% polyacrylamide gels by the method outlined by Laemmli (17). Protein bands in the polyacrylamide gels were visualized by staining with Coomassie brilliant blue R-250 (Bio-Rad, Hercules, Calif.). N-terminal sequencing. N-terminal amino acid sequencing was done by the University of Illinois Biotechnology Center with an Applied Biosystems model 477A protein sequencer with a model on-line PTH analyzer by using Edman degradation chemistry (10). Nucleotide sequence and accession number. The DNA sequence reported here has been submitted to the GenBank database and has been assigned the accession number U82255.

RESULTS AND DISCUSSION Cloning and sequencing of manA. Six thousand plaques were screened for endoglucanase activity by using a chromogenic substrate, OBR-HEC. Eight recombinant phages exhibiting endoglucanase activity were detected and purified. The phagemids isolated from these phages were further analyzed by using endoglucanase assays at both 37 and 65°C. Two restriction endonucleases, EcoRI and PstI, were used to digest the phagemids to examine the restriction patterns of the DNA inserts. The results indicated that seven recombinant phagemids harbored the same segment of the T. polysaccharolyticum genomic DNA in one orientation, while one contained an identical fragment in the opposite orientation. One of the seven similar recombinant phagemids was selected and used for further analysis. This phagemid harbored a DNA insert of approximately 3.4 kb (Fig. 1), which was nucleotide sequenced. The results revealed a large open reading frame (ORF) lacking a termination codon, which indicated an incomplete gene sequence. Using a genome-walking PCR method, an overlapping segment (1.5 kb) of T. polysaccharolyticum genomic DNA was amplified and cloned into a pGEM-T plasmid and nucleotide sequenced to obtain the entire gene. The integrity of the cloned PCR fragment was confirmed through Southern hybridization by using a 32P-labeled fragment amplified from the cloned DNA insert (data not shown). DNA sequence analysis of the manA gene. The cloned gene is designated manA because of its unusually high specific activity on galactomannan (locust bean gum). Figure 1a shows a schematic map of the manA gene. The gene is preceded by the

VOL. 181, 1999

ManA OF T. POLYSACCHAROLYTICUM

1645

FIG. 1. Restriction map and domain organization of ManA and its derivatives, ManA-SL and ManA-SL-CBD. The full length of ManA contains the signal peptide and three other domains. CBD1 and CBD2 are not separated by a Pro-Ser-Thr-rich linker. Sp, signal peptide. The putative mannanase domain is also the catalytic domain. Note that all proteins from T. polysaccharolyticum used in the biochemical analyses have the signal peptide processed.

carboxy terminus of a truncated protein showing 54 and 53% homology to a 51-amino-acid stretch of b-fructofuranosidases from Beta vulgaris (33) and Vicia faba (39), respectively. The start codon, ATG, of manA is located at position 586, which is 22 nucleotides after the termination codon of the truncated ORF, whereas the termination codon, a TAG, is at position 3,877 (Fig. 2). The entire ORF is therefore composed of 3,291 bases and codes for a preprotein of 1,097 amino acid residues with an estimated molecular mass of 119,672 Da. The start codon is preceded by a putative ribosome binding site (TAA GGCGGTG) nine bases upstream and a putative 210 (TAA AAT) and 235 (TTCGC) promoter region occurring within the truncated protein. With the exception of one base, the putative ribosome binding site is a perfect complement of a sequence found at the 39 end of E. coli 16S ribosomal RNA (39-AUUCCUCCAC-59) which plays a crucial role in bringing the 30S ribosome to the initiator codon. Furthermore, an ATrich spacer region is found between the putative Shine-Dalgarno sequence and the initiator (ATG), which is also followed immediately by the sequence AAAA. The requirements for optimal transcription in E. coli (9) are fulfilled by the putative promoter region of the manA gene; hence, the gene is constitutively expressed in E. coli. The production of enzyme from plasmids harboring the insert in both orientations indicated that the promoter sequence of manA was recognized by the E. coli transcriptional apparatus. The G1C content of the region upstream of manA was 51.3%, whereas the region coding for ManA was 46.2%, which reflected the overall G1C content of T. polysaccharolyticum genomic DNA (46%). On the other hand, the region which does not contain an ORF located immediately downstream of manA had a G1C content of 39.7%. Domain analysis of ManA. The N-terminal amino acid sequence of ManA has the features of a typical signal peptide (Fig. 2), a characteristic of extracellular enzymes. When the N-terminal region of ManA is compared with the cleavage sites of other signal peptides, there appears to be a typical signal peptidase I Ala-X-Ala processing site at amino acid residue 32 (22). The N-terminal amino acid sequence of recombinant ManA purified from E. coli was AGTSGDGRFHV, which

corresponds to the amino acid sequence from positions 33 to 43 and suggests that this protein is secreted by both E. coli and T. polysaccharolyticum. The residues of the signal peptide of ManA show high homology to those of several proteins in the GenBank, notably 77 and 64% homology to the major outer sheath signal peptide of Treponema denticola (11) and that of the b-lactamase of Lactococcus lactis (35), respectively. Computer-assisted homology analyses with the National Library of Medicine retrieval system (http://www.ncbi.nlm.nih .gov/) and the BLAST algorithm (2) to scan GenBank and other nonredundant databases indicated that ManA is a modular protein comprising four domains. In Fig. 1b, the sequence of the first 174 amino acids of the region designated as a catalytic domain (amino acid residues 39 to 212; Fig. 2) shows some homology to endoglucanases belonging to cellulase subfamily A5. The amino acid residues from this region of ManA aligned with members of this family (Fig. 3), however, show only a few conserved amino acids in ManA. In this region, the most obvious feature is the conserved amino acid sequence, NEP, which has been shown to be the catalytic center of members of this subfamily. Mutation of the glutamic acid (E) to alanine (A) abolished enzymatic activity (30). This region of ManA showed the highest homologies to CelA of Butyrivibrio fibrisolvens (22.8% identity [15]) and EglA of Clostridium acetobutylicum (20.2% identity [43]). The sequence of the next 424 amino acids (amino acids 213 to 636) shows very little homology to sequences in the public databases (GenBank, EMBL, DDBJ, and Swissprot) except for a stretch of 53 amino acid residues (residues 240 to 292; Fig. 2) which shows homology to the endo-1,4-b-mannosidases of C. saccharolyticum (56% [14]), and Streptomyces lividans (56% [4]). The similarity of the initial amino acid sequence of the catalytic domain to those of the members of cellulase family A5, together with the putative active site, NEP, seemed to indicate that this domain comprised two separate catalytic domains, one for cellulase and the other for mannanase activities. In polysaccharide depolymerases, when two different catalytic domains are present in the same protein, they are usually separated by a linker. However, we did not observe a classical Pro-Ser-Thr-rich (PT box) linker in ManA as has been ob-

1646

CANN ET AL.

J. BACTERIOL.

FIG. 2. Nucleotide and amino acid sequences of manA, the mannanase gene of T. polysaccharolyticum. The locations of the 235 and 210 sites and ribosome binding sites are shown. The sequences of the signal peptide ( ) the CBDs ( ), and SLH ( ) domains are also demarcated. The asterisk shows the stop codon.

12

served in the domain boundaries of the mannanase of C. saccharolyticum (14) and other cellulolytic and hemicellulolytic enzymes, except for the following sequence: PVTSPELTVSKT DFTVKAVIDSSST (positions 441 to 465). Competition studies with both CMC and locust bean gum on OBR-HEC sug-

gests that both cellulase and mannanase activities evolve from the same active site (data not shown). Following the catalytic domain is a duplicated region of 120 highly conserved amino acids, cellulose binding domain 1 (CBD1) and CBD2. The two regions, which are separated by

ManA OF T. POLYSACCHAROLYTICUM

VOL. 181, 1999

1647

FIG. 2—Continued.

23 amino acids (Fig. 2), exhibit 69% identity. The 23-aminoacid spacer sequence shows homology to regions without any assigned function in several proteins in the GenBank database, notably 81 and 61% homology to sequences found at the C terminus of an endo-1,3(4)-b-glucanase of Clostridium thermocellum (accession number X89732) and the central region of a xylanase of Prevotella ruminicola (accession number Z79595), respectively. On the other hand, the two conserved repeats show no meaningful homology to any sequence in both the GenBank and Swissprot databases, except for some regions of

a wall-associated protein precursor of Bacillus subtilis (42) and a transferrin-binding protein of Neisseria gonorrhoeae (3). Through comparisons with the architecture of other polysaccharide depolymerases, this domain was predicted to be a CBD. To test this hypothesis, a crude extract of the protein was applied to a cellulose column, followed by extensive washing with 50 mM sodium phosphate buffer (pH 5.8) while monitoring the protein elution. Desorption of CBDs from cellulose is normally achieved by the use of water, organic solvents, or denaturants (guanidium hydrochloride, urea, or SDS). Our

FIG. 3. Alignment of family A5 domain of T. polysaccharolyticum ManA, B. subtilis DLG End2 (29), C. acetobutylicum Egl (43), B. subtilis CK2 Cel (21), B. subtilis endo-b-1,4-glucanase (32), B. fibrisolvens CelA (15), and Bacillus sp. strain CelB (34). The boldface letters show the conserved residues containing the catalytic center of cellulase family A5. The asterisks indicate amino acid conservation in all enzymes compared. Adjacent amino acids are separated by dashes, where necessary, for optimal alignment. All sequences are aligned from Met-1 of the peptide.

1648

CANN ET AL.

J. BACTERIOL.

FIG. 4. Multiple sequence alignment of the SLH domains of T. polysaccharolyticum ManA (TpolmanA), Thermoanaerobacterium endoxylanase (TmSpXyn [22]), T. thermosulfurigenes pullulanase (TmTamypul [26]), T. saccharolyticum endoxylanase (TmSxyn [18]), and C. thermocellum exoglucanase (CITexoglu; accession number P38535). SLH-1, SLH-2, and SLH-3 show the triplication of this domain. The asterisks indicate conservation. Numbering for each protein starts from the putative initiation codons.

first elution trial was with water, which successfully eluted two proteins with molecular masses of approximately 116 and ,97 kDa (see Fig. 5). The mechanism of desorption is not fully understood, but the conservation of aromatic residues, especially tyrosine and tryptophan, are considered to be crucial for binding (37). Within each of the two repeats in ManA, there are five and three conserved residues of tyrosine and tryptophan, respectively (Fig. 2). The smaller protein band (see Fig. 5, lane 4), which from our molecular mass estimation lacks the SLH repeats and all or part of CBD2 (Fig. 1c, ManA-SL), binds to cellulose and also possesses mannanase activity (see Fig. 6). From this observation, it could be inferred that CBD1 may be enough to confer the cellulose binding property to ManA. To confirm the cellulose binding property of the 120amino-acid repeats, a PCR method was used in deleting both sequences (CBD1 and CBD2) together with the S-layer repeats. The truncated protein (ManA-SL-CBD; Fig. 1d) when applied to the cellulose column failed to bind to cellulose. In addition, the deletion resulted in a negligible hydrolysis of both locust bean gum (galactomannan) and CMC. This observation is similar to results obtained from the CelG of Clostridium cellulolyticum, where the deletion of its putative family III CBD abolished enzymatic activity (12). On the contrary, the removal of the CBD of B. succinogenes endoglucanase by proteolysis decreased the activity of the enzyme twofold on Avicel, without a significant change of activity on CMC (27). A number of polysaccharide depolymerases have been found to contain CBDs which are thought to enhance enzymatic activity on insoluble substrates by increasing the effective enzyme concentration on the substrate surface (37) and also by disrupting the structure and hence increasing accessibility (5). The significant loss of both mannanase and carboxy methylcellulase (CMCase) activities when CBD1 and CBD2 were deleted indicated the importance of this region to the structure and function of this modular protein. Perhaps, the CBDs of ManA, in addition to facilitating the binding of substrate, also possess substrate disruption properties. The results also seem to suggest that at least one of the CBDs is required for the proper folding of this protein. Recently, CBDs were classified into nine families (37) according to amino acid sequence homology. The amino acid sequence of the CBDs of T. polysaccharolyticum ManA shows

no similarity to any known CBD; hence, the sequence describes a novel CBD. The carboxy terminus of ManA contains three SLH repeats of 59 (SLH-1), 67 (SLH-2), and 56 (SLH-3) amino acid residues, respectively. Alignment of the triplicated repeats with those described for other enzymes is shown in Fig. 4. The ManA SLH repeats are homologous to those found in xylanases, such as the endoxylanase of Thermoanaerobacterium sp. (22), the pullulanase of Thermoanaerobacterium thermosulfurigenes (26), the endoxylanase of Thermoanaerobacterium saccharolyticum B6A-RI (18), and the exoglucanase of C. thermocellum (accession number P38535). These SLH repeats were originally proposed to serve in anchoring the glucanases to the peptidoglycan (23). The evidence for this interaction was provided by others (19) who, in constructing a chimeric protein comprising the surface layer repeats and the E. coli maltose binding protein (MalE), conferred the ability to bind covalently to the peptidoglycan of C. thermocellum to the MalE protein. To our knowledge, ManA is the only mannanase exhibiting the surface layer homology to be described in the literature. Purification and characterization of manA. In order to clone the complete manA gene in E. coli, primers were designed to amplify the complete ORF of manA by PCR with T. polysaccharolyticum genomic DNA as the template. The recombinant protein produced was purified from E. coli cells harboring the gene by a two-step purification program, a heating step, and a cellulose affinity step (affinity chromatography). Heat treatment of cell extracts increased the activity of ManA 5.2-fold through the denaturation of host proteins. Application of this partially purified protein to a column of Sigmacell followed by elution with water yielded two protein bands migrating to approximately 116 and ,97 kDa on SDS-PAGE (Fig. 5). The 116-kDa protein corresponded well with the molecular mass estimated from the nucleotide sequence of manA. Zymogram analysis indicated that both protein bands have mannanase activity (Fig. 6). The results of N-terminal amino acid sequencing showed that both proteins possess the same N-terminal sequence (AGTSGDGRFHV), suggesting that the smaller size protein (,97-kDa band) was a result of truncation at the carboxy terminus. A derivative of ManA with a processed

VOL. 181, 1999

ManA OF T. POLYSACCHAROLYTICUM

1649

FIG. 5. Purification steps of recombinant ManA from E. coli cells. Proteins were loaded on and SDS–10% polyacrylamide gel and stained with Coomassie brilliant blue. Lanes: 1, Bio-Rad molecular mass markers; 2, total cell extract; 3, supernatant after heat treatment at a 70°C for 15 min; 4, purified ManA and a truncated derivative (ManA-SL) eluted from a Sigmacell column with water. Molecular mass markers are indicated on the left in kilodaltons.

signal peptide and the SLH domains deleted will be approximately 95 kDa in mass. It is our estimation that the truncated protein lacks the C-terminal S-layer repeats and probably also a segment of CBD2. The ability to bind to cellulose and the possession of enzymatic activity confirmed our assumption that the SLH repeats are not required for binding to substrate and enzymatic activity. ManA activity exhibited a narrow pH range (Fig. 7a), with the optimum pH occurring around 5.8, which reflected the acidic conditions of the site from which T. polysaccharolyticum was isolated. With CMC as the substrate, ManA exhibited activity across a broad temperature range. We observed two temperature optima for CMCase activity (Fig. 7b), which is unusual. One optimum was at 65°C, which coincided with the optimum temperature for growth of T. polysaccharolyticum, and the other was ca. 75°C, which is just above the optimum temperature range of mannanase activity (Fig. 7b). ManA is a multidomain protein, and a long stretch of this protein lacks significant amino acid sequence similarity to any known protein. The number of domains we have assigned to ManA is based on biochemical analyses and sequence similarities to proteins found in publicly available databases (GenBank, Swissprot, and DDBJ). Hence, we cannot overrule the possibility of a catalytic domain eluding us due to insufficient information. Consequently, the presence of a second optimum temperature for CMCase activity may be due to another catalytic domain which is activated at 75°C. On the contrary, this observation could be due to a conformational change in the

FIG. 6. A zymogram showing hydrolysis of locust bean gum (galactomannan) by both proteins eluted from a Sigmacell column (Fig. 5, lane 4). Equal amounts of eluted proteins were loaded in each lane (1 to 4) to show the consistency of hydrolysis.

FIG. 7. Effect of pH and temperature on the hydrolysis of substrates by ManA. (a) Enzyme assays were carried out at 65°C for 10 min in 50 mM sodium phosphate buffer at pH 4.5 to 8.0. Reaction mixtures (total volume, 500 ml) contained 450 ml of 0.5% locust bean gum and 50 ml of enzyme. The final concentration of enzyme was 200 ng/ml, and the pH was measured at 65°C. (b) Enzyme activity was assayed in 50 mM phosphate buffer at pH 5.7 at 40 to 80°C. The final concentration of enzyme in the assay for mannanase (h) and CMCase (F) activities were 200 ng/ml and 1.8 mg/ml, respectively. Relative activity was defined as a proportion of the highest absorbance at 410 nm.

overall structure of ManA which favors CMC hydrolysis at ca. 75°C. Accumulation of more sequences and further experimentation are likely to provide an explanation for this finding. The specific activities of ManA with locust bean gum, lichenan, CMC, and xylan as substrates at 65°C were 1,412, 456, 167, and 10 U/mg, respectively. Laminarin and chitin were not hydrolyzed by ManA. From our analyses it seems that ManA is a multifunctional enzyme tethered to the surface of T. polysaccharolyticum. Its close location is advantageous for the organism because of the release of its soluble hydrolytic products in close proximity to the cells. This would facilitate transport into the cell for me-

1650

CANN ET AL.

J. BACTERIOL.

tabolism. Indeed, sequence analysis of the truncated gene immediately upstream of manA suggests that this ORF is preceded by a putative transporter (.90% identity to a membrane component of an ABC transporter of T. thermosulfurigenes; accession number U50952) which is probably involved in the transport of monomers and short oligomers derived from the hydrolysis of polysaccharides by ManA. This hypothesis, however, needs experimental verification. A great diversity of polysaccharides abound in nature due to the existence of a wide variety of monosaccharides. Consequently, numerous polysaccharide-hydrolyzing enzymes have evolved in various organisms. Whereas some bacteria produce several enzymes to accomplish the task of substrate hydrolysis, others, especially the thermophilic bacteria, in an attempt to increase efficiency, have evolved enzymes combining several functions or catalytic activities on a single protein tethered to the cell surface. The organization of ManA, as described here, is yet further evidence for the versatility of bacteria when faced with the challenge involved in plant cell wall degradation. ACKNOWLEDGMENT This work was supported by a grant from the Illinois Council for Agricultural Research (CFAR-Illinois). REFERENCES 1. Ahsan, M. M., T. Kimura, T. Karita, S. Karita, and K. Ohmiya. 1996. Cloning, DNA sequencing, and expression of the gene encoding Clostridium thermocellum cellulase CelJ, the largest catalytic component of the cellulosome. J. Bacteriol. 178:5732–5740. 2. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410. 3. Anderson, J. E., P. F. Sparling, and C. N. Cornelissen. 1994. Gonococcal transferrin binding protein 2 facilitates but is not essential for transferrin utilization. J. Bacteriol. 176:3162–3170. 4. Arcand, N., D. Kluepfel, F. W. Paradis, R. Morosoli, and F. Shareck. 1993. b-Mannanase of Streptomyces lividans 66: cloning and DNA sequence of the manA gene and characterization of the enzyme. Biochem. J. 290:857–863. 5. Beguin, P., and J.-P. Aubert. 1994. The biological degradation of cellulose. FEMS Microbiol Rev. 13:25–58. 6. Birnboim, H. C., and J. Doly. 1979. A rapid extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513–1523. 7. Buchert, J., J. Salminen, M. Sika-aho, M. Ranua, and L. Viikari. 1993. The role of Trichoderma reesei xylanase and mannanase in the treatment of softwood kraft pulp prior to bleaching. Holzforschung 47:473–478. 8. Cann, I. K. O., P. G. Stroot, K. R. Mackie, B. A. White, and R. I. Mackie. Characterization of two novel saccharolytic anaerobic thermophiles, Thermoanaerobacterium polysaccharolyticum gen. nov., sp. nov., and Thermoanaerobacterium zeae gen. nov., sp. nov. Int. J. Syst. Bacteriol., submitted for publication. 9. Das, A. 1990. Overproduction of proteins in Escherichia coli: vectors, hosts, and strategies, p. 193–112. In M. P. Deutscher (ed.), Guide to protein purification. Academic Press, London, United Kingdom. 10. Edman, P., and G. Begg. 1967. A protein sequenator. Eur. J. Biochem. 1:80–91. 11. Fenno, J. C., G. W. Wong, P. M. Hannam, K. H. Muller, W. K. Leung, and B. C. McBride. 1997. Conservation of msp, the gene encoding the major outer membrane protein of oral Treponema sp. J. Bacteriol. 179:1082–1089. 12. Gal, L., C. Gaudin, A. Belaich, S. Pages, C. Tardiff, and J.-P. Belaich. 1997. CelG from Clostridium cellulolyticum: a multidomain endoglucanase acting efficiently on crystalline cellulose. J. Bacteriol. 179:6595–6601. 13. Flint, H. J., J. Martin, C. A. MacPherson, A. S. Daniel, and J.-X. Zhang. 1993. A bifunctional enzyme, with separate xylanase and b(1,3-1,4)-glucanase domains, encoded by the xynD gene of Ruminococcus flavefaciens. J. Bacteriol. 175:2943–2951. 14. Gibbs, M. D., D. J. Saul, E. Luthi, and P. L. Bergquist. 1992. The b-mannanase from Caldocellum saccharolyticum is part of a multidomain enzyme. Appl. Environ. Microbiol. 58:3864–3867. 15. Hazlewood, G. P., K. Davidson, J. I. Laurie, M. P. Romaniec, and H. J. Gilbert. 1990. Cloning and sequencing of the celA gene encoding endoglucanase A of Butyrivibrio fibrisolvens strain A46. J. Gen. Microbiol. 136:2089– 2097. 16. Higgins, D. G., A. J. Bleasby, and R. Fuchs. 1991. CLUSTAL V: improved software for multiple sequence alignment. Comput. Applic. Biosci. 8:189– 191.

17. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the heat of bacteriophage T4. Nature 227:6809–685. 18. Lee, Y. E., S. E. Lowe, and J. G. Zeikus. 1993. Gene cloning, sequencing, and biochemical characterization of endoxylanase from Thermoanaerobacterium saccharolyticum B6A-RI. Appl. Environ. Microbiol. 59:3134–3137. 19. Lemaire, M, H. Ohayon, P. Gounon, T. Fujino, and P. Beguin. 1995. OlpB, a new outer layer protein of Clostridium thermocellum, and binding of its S-layer-like domains to components of the cell envelope. J Bacteriol. 177: 2451–2459. 20. Lever, M. 1973. Colorimetric and fluorometric carbohydrate determination with p-hydroxybenzoic acid hydrazide. Biochem. Med. 7:274–281. 21. Lindahl, V., K. Aa, and A. Tronsmo. 1994. Nucleotide sequence of an endo-b-1,4-glucanase gene from Bacillus subtilis CK-2. Antonie Leeuwenhoek 66:327–332. 22. Liu, S.-Y, F. C. Gherardini, M. Matuschek, H. Bahl, and J. Wiegel. 1996. Cloning, sequencing, and expression of the gene encoding a large S-layerassociated endoxylanase from Thermoanaerobacterium sp. strain JW/SL-YS 485 in Escherichia coli. J. Bacteriol. 178:1539–1547. 23. Lupas, A., H. Engelhardt, J. Peters, U. Santarius, S. Volker, and W. Baumeister. 1994. Domain structure of the Acetogenium kivui surface layer revealed by electron crystallography and sequence analysis. J. Bacteriol. 176:1224–1233. 24. Luthi, E., N. B. Jasmat, R. A. Grayling, D. R. Love, and P. L. Bergquist. 1991. Cloning, sequence analysis, and expression in Escherichia coli of a gene coding for a b-mannanase from the extremely thermophilic bacterium Caldocellum saccharolyticum. Appl. Environ. Microbiol. 57:694–700. 25. Marmur, J. 1961. A procedure for the isolation of deoxyribonucleic acid from microorganisms. J. Mol. Biol. 3:208–218. 26. Matuschek, M., G. Burchhardt, K. Sahm, and H. Bahl. 1994. Pullulanase of Thermoanaerobacterium thermosulfurigenes EM1 (Clostridium thermosulfurogenes): molecular analysis of the gene, composite structure of the enzyme, and a common model for its attachment to the cell surface. J. Bacteriol. 176:3295–3302. 27. McGavin, M., and C. W. Forsberg. 1989. Catalytic and substrate binding domains of endoglucanase 2 from Bacteroides succinogenes. J. Bacteriol. 171:3310–3315. 28. Morris, D. D., R. A. Reeves, M. D. Gibbs, D. J. Saul, and P. L. Bergquist. 1995. Correction of the b-mannanase domain of the CelC pseudogene from Caldocellulosiruptor saccharolyticus and activity of the gene product on kraft pulp. Appl. Environ. Microbiol. 61:2262–2269. 29. Nakamura, A., T. Uozumi, and T. Beppu. 1987. Nucleotide sequence of a cellulase gene of Bacillus subtilis. Eur. J. Biochem. 164:317–320. 30. Py, B., I. Bortoli-German, J. Haiech, M. Cippaux, and F. Barras. 1991. Cellulase EGZ of Erwinia chrysanthemi: structural organization and importance of His98 and Glu133 residues for catalysis. Protein Eng. 4:325–333. 31. Richmond, P. A. 1991. Biosynthesis and Biodegradation of Cellulose, p. 5–32. In C. H. Haigler, and P. J. Weimer (ed), Marcel Dekker, New York. 32. Robson, L. M., and G. H. Chambliss. 1987. Endo-b-1,4-glucanase gene of Bacillus subtilis. DLG J. Bacteriol. 169:2017–2025. 33. Roitsch, T., M. Bittner, and D. E. Godt. 1995. Induction of apoplastic invertase of Chenopodium rubrum by D-glucose and a glucose analog and tissue specific expression suggest a role in sink-source regulation. Plant Physiol. 108:285–294. 34. Sanchez-Torres, J., P. Perez, and R. I. Santamaria. 1996. A cellulase gene from a new alkalophilic Bacillus sp. (strain N186-1). Its cloning, nucleotide sequence and expression in Escherichia coli. Appl. Microbiol. Biotechnol. 46:149–155. 35. Sibakov, M., T. Koivula, A. Von Wright, and I. Pavla. 1991. Secretion of TEM b-lactamase with signal sequences isolated from the chromosome of Lactococcus lactis. Appl. Environ. Microbiol. 57:341–348. 36. Talbot, G., and J. Sygush. 1990. Purification and characterization of thermostable b-mannanase and a-galactosidase from Bacillus stearothermophilus. Appl. Environ. Microbiol. 56:3505–3510. 37. Tomme, P., R. A. J. Warren, and N. R. Gilkes. 1995. Cellulose hydrolysis by bacteria and fungi. Adv. Microb. Physiol. 37:1–81. 38. Warren, R. A. J. 1996. Microbial hydrolysis of polysaccharides. Annu. Rev. Microbiol. 50:183–212. 39. Weber, H., L. Borisjuk, U. Heim, P. Buchner, and U. Wobus. 1995. Seed coat-associated invertases of fava bean control both unloading and storage functions: cloning of cDNAs and cell type-specific expression. Plant Cell. 7:1835–1846. 40. Wolf, M., G. Attila, S. Ortwin, and B. Rainer. 1995. Genes encoding xylan and b-glucan hydrolysing enzymes in Bacillus subtilis: characterization, mapping and construction of strains deficient in lichenase, cellulase and xylanase. Microbiology 141:281–290. 41. Wong, K. K. Y., and J. N. Saddler. 1993. Applications of hemicellulases in the food, feed, and pulp and paper industries, p. 127–1143. In M. P. Coughlan and G. P. Hazlewood (ed.), Hemicellulose and hemicellulases. Portland Press, Ltd., London, United Kingdom. 42. Yoshida, K., H. Sano, S. Seki, M. Oda, M. Fujimura, and Y. Fujita. 1995. Cloning and sequencing of a 29kb region of the Bacillus subtilis genome containing the hut and WapA loci. Microbiology 14:337–343.

VOL. 181, 1999 43. Zappe, H., W. A. Jones, D. T. Jones, and D. R. Woods. 1988. Structure of an endo-b-1,4-glucanase gene from Clostridium acetobutylicum P262 showing homology with endoglucanase genes from Bacillus sp. Appl. Environ. Microbiol. 54:1289–1292. 44. Zhang, J.-X., and H. J. Flint. 1992. A bifunctional xylanase encoded by the xynA gene of the rumen cellulolytic bacterium Ruminococcus flavefaciens 17

ManA OF T. POLYSACCHAROLYTICUM

1651

comprises two dissimilar domains linked by an asparagine/glutamine-rich sequence. Mol. Microbiol. 6:1013–1023. 45. Zverlov, V., S. Mahr, K. Reidel, and K. Bronnenmeier. 1998. Properties and gene structure of a bifunctional cellulolytic enzyme (CelA) from the the extreme thermophile Anaerocellum thermophilum with separate glycosyl hydrolase family 9 and 48 catalytic domains. Microbiology 144:457–465.