The binding pattern of two carbohydrate-binding modules of laminarinase Lam16A from Thermotoga neapolitana: differences in ß-glucan binding within family CBM4

Vladimir V. Zverlov1, Ilia Y. Volkov1, Galina A. Velikodvorskaya1 and Wolfgang H. Schwarz2

Institute of Molecular Genetics, Russian Academy of Science, Kurchatov Sq., 123182 Moscow, Russia1
Research Group Microbial Biotechnology, Technische Universität München, Am Hochanger 4, D-85350 Freising-Weihenstephan, Germany2

Author for correspondence: Wolfgang H. Schwarz. Tel: +49 8161 71 5445. Fax: +49 8161 71 5475. e-mail: schwarz{at}mikro.biologie.tu-muenchen.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Carbohydrate-binding modules (CBMs) are often part of the complex hydrolytic extracellular enzymes from bacteria and may modulate their catalytic activity. The thermostable catalytic domain of laminarinase Lam16A from Thermotoga neapolitana (glycosyl hydrolase family 16) is flanked by two CBMs, 148 and 161 aa long. They share a sequence identity of 30%, are homologous to family CBM4 and are thus called CBM4-1 and CBM4-2 respectively. Recombinant Lam16A proteins deleted for one or both binding modules and the isolated module CBM4-1 were characterized. Proteins containing the N-terminal module CBM4-1 bound to the soluble polysaccharides laminarin (1,3-ß-glucan) and barley 1,3/1,4-ß-glucan, and proteins containing the C-terminal module CBM4-2 bound additionally to curdlan (1,3-ß-glucan) and pustulan (1,6-ß-glucan), and to insoluble yeast cell wall ß-glucan. The activity of the catalytic domain on soluble 1,3-ß-glucans was stimulated by the presence of CBM4-1, whereas the presence of CBM4-2 enhanced the Lam16A activity towards gelatinized and insoluble or mixed-linkage 1,3-ß-glucan. Thermostability of the catalytic domain was not affected by the truncations. Members of family CBM4 can be divided into four subfamilies, members of which show different polysaccharide-binding specificities corresponding to the catalytic specificities of the associated hydrolytic domains.

Keywords: Thermotoga neapolitana, 1,3-ß-glucanase, substrate-binding module CBM4, binding specificity

Abbreviations: CBM, carbohydrate-binding module; GHF, glycosyl hydrolase family


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Polysaccharides containing 1,3-ß-linkages are abundant in biomass and widespread in nature, especially in marine environments. They play an important role in the structural stability of eukaryotic cell walls, e.g. as a main cell wall component in yeast, filamentous fungi and algae, and have the ability to enhance and stimulate the human immune system (Kulicke et al., 1997 ). Many of the 1,3-ß-glucans are insoluble due to the regularity of their 1,3-ß-linked structure or are gelatinizing in different forms. Some are readily soluble, like the mixed-linkage glucans, due to the irregularly alternating 1,3-ß and 1,4-ß bonds (Harada, 1992 ; Jiang & Vasanthan, 2000 ; Kim et al., 2000 ).

Soluble 1,3-ß-glucans are readily degraded by endoglucanases secreted by a number of bacteria (Schwarz et al., 1988 ; Nogi & Horikoshi, 1990 ; Aida et al., 1995 ; Zverlov et al., 1997 ; Krah et al., 1998 ). The enzymic hydrolysis of insoluble and gelatinized ß-glucans, which are protected by their highly ordered structure, is more problematic (Kanzawa et al., 1994 ) and may be enhanced if hydrolytic enzyme modules are sterically correctly connected with carbohydrate-binding modules (CBMs). These non-catalytic modules were originally detected as ‘cellulose-binding domains’ binding to crystalline and amorphous cellulose (Tomme et al., 1995 ). The development of techniques to study the binding to soluble substrates stimulated the research of binding modules greatly (Tomme et al., 1996 ). They were found to be present in many modular bacterial polysaccharide hydrolases as small, separately folding, non-catalytic protein domains or modules. Twenty-four CBM families have been defined by comparing their primary structure (Tomme et al., 1998 ; Meissner et al., 2000 ; Coutinho & Henrissat, 1999 ). Members of the same CBM family can be connected to catalytic modules of different glycosyl hydrolase families (GHFs) and some modular enzymes contain more than one CBM and also CBMs of more than one family. The nomenclature of CBMs that has been proposed recently by Henrissat et al. (1998) will be used here.

The three-dimensional structure of a number of CBMs has been resolved, most of them having ß-sheet topologies (reviewed by Linder & Teeri, 1997 ; Bayer et al., 1998 ). CBM1, CBM2, CBM3 and CBM5 modules have a planar hydrophobic surface, on which substrate binding is mediated by interaction between exposed aromatic amino acid residues and the glucosyl-pyranose ring. These CBMs interact with multiple cellulose chains and strongly prefer insoluble microfibrils of cellulose to soluble polysaccharide molecules (Bolam et al., 1998 ). In contrast, the CBM4 module CfCenC/N1 has a binding groove that recognizes one polysaccharide strand (Tomme et al., 1998 ). However, the majority of CBMs are not functionally characterized as yet. The binding pattern of CBMs that have been determined can vary widely, even within one family.

The presence of these modules may influence the substrate range of their hydrolytic partner modules, especially on complex substrates like crystalline cellulose (Irwin et al., 1998 ; Gill et al., 1999 ) and complex hemicellulose (Black et al., 1996 ). Removal of the CBMs from a modular protein often reduces the catalytic activity on insoluble glycans without affecting the activity on soluble substrates. The activity enhancement can be explained by an increase in the effective concentration of the hydrolytic enzyme on the macromolecular substrate, either insoluble, partially soluble, gelatinized or bound to an insoluble substrate, as would be the case with hemicellulose bound to the cellulose matrix in plant cell walls (Black et al., 1996 ). Moreover, it has to be pointed out that the degradation of crystalline substrates, besides the hydrolysis of the glycosidic bond, has to accomplish the dislocation of the attacked molecule from the highly ordered, crystalline surface, a (non-catalytic) activity which could be ascribed to the strongly binding CBMs. The CBMs in some enzymes have the ability to thermally stabilize the attached catalytic domain (Meissner et al., 2000 ; Sunna et al., 2000 ), although this seems to be a speciality of co-evolved modules, and a general thermostabilizing role can not be ascribed to them.

Thermotoga neapolitana, like its close relative Thermotoga maritima, is a saccharolytic, Gram-negative bacterium which was isolated from marine environments with volcanic activity. It is, with a maximum growth temperature of 90 °C, the most thermophilic saccharolytic bacterium isolated so far (Huber & Stetter, 1992 ) and constitutes one of the deepest branches in the phylogenetic tree of the bacteria. It hydrolyses many polysaccharides, including 1,3-ß-glucan, with a range of extracellular enzymes (Dakhova et al., 1993 ; Bronnenmeier et al., 1995 ; Zverlov et al., 1999 ) which, due to their inherent stability, are of utmost interest for biotechnological applications. The DNA sequence of the thermostable 1,3-ß-glucanase (laminarinase) gene lamA of T. neapolitana has been determined previously (Zverlov et al., 1997 ). The protein has a modular structure and consists of catalytic and non-catalytic modules. The present study investigates the role of the non-catalytic modules of Lam16A (formerly designated LamA) in the binding to soluble and insoluble substrate molecules. The primary structure of the Lam16A CBMs was compared to known CBMs and to the sequence database. Homologous sequences belong to family CBM4 and obviously fall into four subfamilies. The binding pattern of at least one member of each subfamily was determined or known. The data suggest a correlation with the hydrolytic activity of the catalytic module connected to the CBM.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
Escherichia coli strain XL-1 Blue was used for cloning and strain M15 containing plasmid pREP-4 (Qiagen) for overexpression of cloned genes. Cultivation of recombinant cells, media and overexpression were done as recommended by the manufacturers (Qiagen). Plasmids pQE32 and pQE30 were used for cloning of the PCR products obtained with primers pBamI-1 and pBamI-2 respectively.

Molecular biological methods.
PCR was carried out using oligonucleotide primer pairs with chromosomal DNA as the template and the Expand High Fidelity PCR System (Boehringer Mannheim). The oligonucleotide primers were: for the amplification of LamA-H1 primer pBamI-1 (5'-TGTTTGGATCCTGGCTCAGAATATTTTAC-3') and pSalI-c3 (5'-CTTATGTCGACTCTCGATCTCCCCTCTTC-3'); for LamA-H2, pBamI-1 and pSalI-c2 (5'-TGATCGTCGACTATCGGTTCATCGAATGTG-3'); for LamA-H3, pBamI-2 (5'-ATGTGGGATCCCTGGAAGTCAGTGGTGAGG-3') and pSalI-c3; for LamA-H4, pBamI-2 and pSalI-c2; and for LamA-H5, pBamI-1 and pSalI-c1 (5'-GACAAGTCGACAGTCTTCCACCTTGTCCTC-3').

The DNA sequence of cloned PCR products was verified by sequencing supercoiled double-stranded plasmid DNA on both strands by using the Thermo Sequenase Cycle Sequencing Kit (Amersham) for extension of 5'-biotinylated primers. DNA fragments were separated with a GATC 1500 Direct-Blotting-Electrophoresis apparatus and detected using streptavidin-conjugated alkaline phosphatase and nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (Serva) as the chromogenic substrate. Sequence data were analysed with the DNASIS/PROSIS software package (Hitachi Software Engineering). Nucleotide and protein databases were screened using the FASTA and BLAST software at the EBI-server at EMBL (http://www.ebi.ac.uk). Sequence alignment was done with the CLUSTAL W software at the same server.

Purification of recombinant proteins.
Recombinant proteins were purified from 50 ml IPTG-induced E. coli cultures with Ni-NTA spin columns according to the manufacturer’s instructions (Qiagen). The purity of the proteins was verified by SDS-PAGE and staining with Coomassie brilliant blue G-250 (Serva).

Enzyme assays.
Enzyme aliquots in standard assays were incubated in phosphate/citrate buffer (50 mM, pH 6·2) with 0·7% (w/v) substrates at 85 °C. Reducing sugars released from polymeric substrates were detected by the 3,5-dinitrosalicylic acid method (Wood & Bhat, 1988 ), assuming that 1 U enzyme liberates 1 µmol glucose equivalent min-1 (mg protein)-1. To get values within the linear rate of reaction, the amount of enzyme and the time of incubation were chosen to result in values well below the saturation point of the test. All tests were performed in triplicate and the coefficients of variation (standard deviation divided by the mean) were less than 8%.

Analysis of binding to soluble substrates by affinity electrophoresis.
The affinity electrophoresis method described by Tomme et al. (1996) was applied. Native polyacrylamide gels (10%), with and without binding substrates (1 mg ml-1) added prior to polymerization, were prepared separately and run in parallel at 4 °C. BSA fraction V was used as a negative non-interacting control. Proteins were visualized by staining with Coomassie brilliant blue G-250.

The retardation coefficient Kr was quantified by calculating the relative migration distances r (with substrate) and r0 (without substrate): Kr=rx(r0-r)-1x[S], where [S] is the substrate concentration (mg ml-1). The relative migration distance, r, was determined as the ratio of the migration distance of the major protein band and the migration front of the gel.

Analysis of binding to insoluble substrates.
Binding of proteins to insoluble carbohydrates was carried out in 1·5 ml polypropylene tubes at 4 °C. Purified protein (5–10 µg) was added to 10 mg substrate in 0·5 ml 80 mM imidazole/HCl buffer (pH 6·8) containing 0·2 M NaCl and incubated on a shaker. The samples were then centrifuged for 5 min to remove the ligand together with the bound protein. Free protein in the cleared supernatants was determined with the Protein Assay Kit (Bio-Rad). The measurements were done at least in triplicate using BSA as a negative control. The residual protein concentration (percentage) was determined as the ratio between the initial and the free protein concentration after incubation with the substrate.

Substrates.
Birchwood xylan, chitin from crab shells, avicel CF1, carboxymethylcellulose (low viscosity; degree of substitution, 0·9), lichenan and laminarin were obtained from Sigma-Aldrich, barley ß-glucan and pachyman from Megazym, curdlan from Wako Pure Chemical Industries, and pustulan from Roth. Hydroxyethyl-cellulose HEC20 was manufactured by Clarian as Tylose H20g4 (degree of substitution, 1·2; molar substitution, 1·7) and obtained through Serva. The ß-glucans used in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. The polysaccharides used as substrates in this study

 
Insoluble ß-glucan from yeast cell walls was prepared according to Bacon et al. (1969) from Saccharomyces cerevisiae cells: briefly, 50 g pressed yeast cells were homogenized in 300 ml NaOH (3%, w/v), incubated for 1 h at 75 °C and centrifuged. With the pellet the procedure was repeated four times. The residual pellet was successively washed with phosphate buffer (pH 7·4) and DMSO/H2O (1:1), and dried. It was resuspended in 30 ml DMSO/H2O (9:1) and sonicated. After centrifugation the pellet was washed twice with water and ethanol, and dried.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Sequence analysis of Lam16A and construction of truncated proteins
The nucleotide sequence (3786 bases) of clone pTT26 containing the complete reading frame of T. neapolitana endo-1,3-ß-glucanase lamA has previously been described (Zverlov et al., 1997 ; GenBank accession no. Z47974). Sequence analysis disclosed the modular structure of Lam16A: a central catalytic module belonging to GHF16 was flanked by two non-catalytic modules, designated CBM4-1 and CBM4-2. The non-catalytic modules share 30% identical amino acid residues (42% similarity) and can be assigned to family CBM4 by sequence alignment (Tomme et al., 1998 ). The greatest sequence identity was found with the binding modules of the lamA gene from T. maritima and with the four binding modules of the 1,3-ß-glucanase LicA of Clostridium thermocellum (Fig. 1). Other CBM4 modules, like the well characterized N1 module of CenC from Cellulomonas fimi (now designated CfCel9B/CBM4-1) were less, but still clearly related. According to the systematic nomenclature proposed recently, the Lam16A protein is called TnLam16A and has the structure CBM4-1/CD16/CBM4-2. The single CBMs are unequivocally referred to as TnLam16A/CBM4-1 and TnLam16A/CBM4-2 (Henrissat et al., 1998 ).



View larger version (119K):
[in this window]
[in a new window]
 
Fig. 1. Alignment of binding modules of family CBM4 with CLUSTAL W (manually corrected). Shaded boxes indicate the areas determined as {alpha}-helices and ß-sheets by structural analysis of the Cel. fimi CenC(1) module. Bold letters are amino acid residues essential in Cel. fimi CenC(1): *, essential for binding; #, residues involved in calcium binding; ^, cysteine residues essential for structural integrity. The sequences are as follows: CtLicA, Clo. thermocellum 1,3-ß-endoglucanase LicA (accession no. CAA61884); TnLamA, T. neapolitana 1,3-ß-endoglucanase LamA (CAA88008); TmLamA, T. maritima 1,3-ß-endoglucanase LamA (pir B72428); MxCelA, Myxococcus xanthus 1,4-ß-endoglucanase CelA (CAA54096); TfCelE, Thermobifida fusca 1,4-ß-endoglucanase CelE (AAC06387); SrCel, Streptomyces reticuli 1,4-ß-endoglucanase Cel1 (sp Q05156); ScCel, Streptomyces coelicolor 1,4-ß-endoglucanase Cel1 (CAB61539); CfCenC, Cellulomonas fimi 1,4-ß-endoglucanase CenC (sp P14090); RmXyn, Rhodothermus marinus endoxylanase (CAA72323); RaEng7, Ruminococcus albus endoxylanase EgVII (BAA92430); CtCelK, Clostridium thermocellum cellobiohydrolase CelK (AAC06139); CtCbhA, Clo. thermocellum cellobiohydrolase CbhA (CAA56918); CcCelE, Clostridium cellulolyticum ‘endoglucanase’ CelE (AAA73869); CvEngN, Clostridium cellulovorans ‘endoglucanase’ EngN (AAF61310); CvEngK, Clo. cellulovorans ‘endoglucanase’ EngK (AAF06107); CvEngM, Clo. cellulovorans ‘endogucanase’ EngM, (AAF06111).

 
To experimentally determine the function of non-catalytic modules CBM4-1 and CBM4-2, deletion mutants of lamA were constructed by a PCR-based method. All clones were verified by nucleotide sequencing. The proteins were produced in recombinant E. coli cells from an expression vector introducing a synthetic His6 tag sequence. This allowed for overexpressing the recombinant protein upon induction and for a one-step purification of the recombinant proteins from the cell extracts. The structures of the recombinant proteins are shown in Fig. 2. All recombinant proteins lack the leader peptide. LamA-H1 is the ‘complete’ reference enzyme (Mr 73 kDa), LamA-H2 (Mr 57 kDa) and LamA-H3 (Mr 53 kDa) are missing either CBM4-1 or CBM4-2, LamA-H4 contains only the catalytic domain (Mr 38 kDa), and LamA-H5 only the N-terminal non-catalytic module CBM4-1 (Mr 25 kDa). Protein fractions homogeneous in denaturing SDS-PAGE were obtained and used for the following characterization.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 2. Structure of the laminarinase Lam16A of T. neapolitana and its truncated forms. Hatched boxes indicate binding modules; the white box represents the catalytic domain. LP, leader peptide. The scale at the top of the figure indicates amino acid residues.

 
Polysaccharide binding
The polysaccharide-binding capability of the non-catalytic modules was determined in two ways: by visualizing the adsorption to soluble substrates using gel retardation in native polyacrylamide gel electrophoresis containing the binding substrates (affinity gel electrophoresis) and by assaying the remaining protein in the supernatant after incubation with insoluble binding substrates. In affinity gel electrophoresis the purified catalytic domain, LamA-H4, was not retarded by any soluble substrate, whereas all other constructs were retarded by laminarin and barley ß-glucan, both soluble and 1,3-ß-linkage-containing substrates (Table 2). Fig. 3 is an example of a gel retardation experiment.


View this table:
[in this window]
[in a new window]
 
Table 2. Binding of the recombinant proteins to soluble substrates as determined in gel retardation assays

 


View larger version (65K):
[in this window]
[in a new window]
 
Fig. 3. Retardation of purified Lam16A proteins in native gel electrophoresis containing ß-glucan. (a) Without ß-glucan; (b) with barley ß-glucan. Lanes: M, marker protein (BSA fraction V); 1, LamA-H5; 2, LamA-H4; 3, LamA-H1; 4, LamA-H2; 5, LamA-H3. Gels were run from top to bottom and were stained with Coomassie brilliant blue. Only the major bands were used for calculation. Minor bands constitute contaminating proteins or structurally impaired forms.

 
The retardation coefficient Kr was calculated by comparing the migration distance without and with binding substrate. Kr is analogous to the dissociation constant Kd for insoluble substrates. Proteins were more retarded by ß-glucan than by laminarin, especially if CBM4-2 was present. In addition, the proteins containing the CBM4-2 module (LamA-H1 and -H3) bound to the 1,3-ß-glucan curdlan and the 1,6-ß-glucan pustulan. No retardation was observed with carboxymethylcellulose and xylan, which are 1,4-ß-linked glycans (data not shown). In contrast to these 1,4-ß-linked polymers, hydroxyethylcellulose was efficiently bound by the CBM4-2 module. This binding can only be explained by non-specific binding to the hydroxyethylester side groups, because no binding was seen to other substances containing 1,4-ß-links, either soluble or insoluble. Thus, CBM4-1 seems to bind specifically to 1,3-ß-linkages in soluble glucans, whereas CBM4-2 has a broad specificity for 1,3-ß-, 1,6-ß- and eventually to other linkages.

Binding to insoluble polysaccharides was assayed by measuring the depletion of protein from cleared supernatant. The ratio of unbound to bound enzyme was plotted against the time of incubation (Fig. 4). No binding was observed to the 1,4-ß-linked polysaccharides Avicel (micro-crystalline cellulose), xylan and chitin. The constructs LamA-H1 and LamA-H3 ({Delta}CBM4-1) bound to the insoluble yeast cell wall ß-1,3-glucan, whereas LamA-H2 ({Delta}CBM4-2) and -H4 ({Delta}CBM4-1 {Delta}CBM4-2) did not. The binding kinetics were fast: maximum binding was reached after 5–10 min. Only LamA-H1 and -H3 contained CBM4-2 (Fig. 2). This clearly indicates that the CBM4-2 module is responsible for the specific physical attachment to insoluble 1,3-ß-glucan.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 4. Binding kinetics of Lam16A and its truncated forms to insoluble yeast cell-wall ß-glucan. The ratio of free to bound enzyme is given. {bullet}, LamA-H1; {blacksquare}, LamA-H2; {blacktriangleup}, LamA-H3; {blacktriangledown}, LamA-H4.

 
Enzymic activity of truncated Lam16A
The hydrolytic activity of the deletion proteins on soluble, gelatinized and insoluble ß-glucans containing 1,3-ß-linkages was determined (Table 3). The 1,3-ß-glucans laminarin, curdlan and pachyman, and the mixed linkage 1,3/1,4-ß-glucans barley ß-glucan and lichenan were substrates for Lam16A. Carboxymethylcellulose, carboxyethylcellulose, xylan, chitin and Avicel were not hydrolysed. To compare the activities between truncation variants of one protein having different molecular masses it was necessary to express the specific activity as units per molecule (U µmol-1). The 31·5 U µmol-1 specific activity determined for LamA-H1 on laminarin corresponded to 431 U (mg protein)-1. The highest activity of all enzyme variants on each substrate was found for the complete protein LamA-H1, which contains both non-catalytic modules in the native order (Table 3). All proteins, with and without binding modules, had the highest activity on the substrate laminarin, a soluble 1,3-ß-glucan containing a low number of 1,6-ß-linked glucose branches with a branching ratio of 0·07 (Kim et al., 2000 ). They were less active on the mixed-linkage and the insoluble 1,3-ß-glucans. In comparison to LamA-H1, LamA-H3 (retaining the CBM4-2 module) had lost more than 70% of its activity on laminarin, but only 12–35% of the activity on the other substrates, whereas LamA-H2 (retaining CBM4-1) had lost 39% versus 61–77%, respectively. This indicated that the deletion of module CBM4-1 has not nearly as strong an effect on the activity towards mixed-linkage and insoluble 1,3-ß-glucans as deletion of CBM4-2. In other words, CBM4-1 was necessary for high activity on laminarin, whilst CBM4-2 stimulates activity on 1,3/1,4-ß-glucans and insoluble 1,3-ß-glucans.


View this table:
[in this window]
[in a new window]
 
Table 3. Specific and relative hydrolytic activity of LamA and its derivatives on polymeric substrates

 
The considerable loss of activity on truncation showed the important role of the non-catalytic modules for the hydrolytic activity of the catalytic domain, which might be due to substrate binding. A reduction in protein stability on truncation was ruled out: the deletion of one or both CBMs did not result in a significant decrease of thermostability of LamA-H4; ~80% (±5%) of laminarinase activity was present with all recombinant proteins after 30 min incubation at 95 °C without substrate.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The sequences of the C- and N-terminally located CBMs in the 1,3-ß-glucanase Lam16A of T. neapolitana, CBM4-1 and CBM4-2, are highly homologous. They belong to the module family CBM4. These non-catalytic domains are true CBMs. They bind specifically to ß-glucans containing 1,3-ß-linkages and not to 1,4-ß-linked glycans like cellulose, xylan or chitin. Despite their similarity, significant differences were observed in the binding pattern, with both CBMs binding to the soluble 1,3-ß-glucan laminarin. Only CBM4-2 was able to fully bind to barley ß-glucan (1,3/1,4-ß-glucan). The difference between the two modules was even more pronounced with gelatinized curdlan (1,3-ß-glucan), the 1,6-ß-glucan pustulan and the insoluble yeast cell wall glucan, which were selectively bound only by the CBM4-2 module. The quantitative binding as expressed by the retardation coefficient Kr was calculated from the affinity gel electrophoresis. Its value may be biased by the intrinsic hydrolytic activity of the proteins constructed: although gels were run at 4 °C to minimize hydrolysis of the substrate, enzymic activity could diminish obvious retardation. This effect may partly explain the lower retardation towards laminarin than to barley ß-glucan.

Deletion of the non-catalytic modules diminished the hydrolytic activity of Lam16A. The greatest effect for the hydrolytic activity on soluble 1,3-ß-glucan was observed by deletion of CBM4-1, whereas on insoluble substrates and mixed-linkage glucan deletion of CBM4-2 had the greatest effect. This indicates a co-operativity of the substrate binding with the catalytic activity of the GHF16 module, supporting the model that binding modules enhance the catalytic activity of an enzyme by increasing its concentration on the substrate. This argument holds true at least for soluble substrates. For insoluble ß-glucans it cannot be ruled out that CBM4-2 interacts with the substrate surface and disrupts the ordered surface structure. This would allow the catalytic domain of the enzyme to access the substrate molecule. However, this argument cannot apply for gelatinized ß-glucans. It should also be noted that a thermostabilizing effect, as was described for non-catalytic modules homologous to CBMs in other bacterial exo-enzymes (Meissner et al., 2000 ; Sunna et al., 2000 ), was not observed with the presence of the CBM4-1 and CBM4-2 modules in Lam16A. Therefore it seems unlikely that the loss of activity is due to diminished stability.

Protein databases were screened for sequences similar to the two TnLam16A binding modules: 22 sequences belonging to module family CBM4 were identified, all occurring in secreted glycosyl hydrolases of bacteria, both mesophilic and thermophilic. The sequences were aligned using the neighbour-joining method in progressive multiple sequence alignment (CLUSTAL W; Thompson et al., 1994 ) (Fig. 1). A similarity tree was constructed (Fig. 5) which showed a clear subdivision of the CBM4 family into four subfamilies.



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 5. Dendrogram derived from a similarity alignment of CBM4 modules. The tree topology was evaluated and corrected according to the results obtained from a multiple sequence alignment applying maximum-parsimony and likelihood (CLUSTAL W) and drawn to scale with PHYLIP. The black bar indicates 10% estimated sequence divergence. Sequence designations are as in Fig. 1. Numbers in parentheses at the end of the module designation indicate the number of a module-repeat within a gene, beginning with the N terminus.

 
All subfamilies (except CBM4D) contain multiple copies of binding modules present in one enzyme. It is obvious that if more than one copy of a CBM4 module is attached to one catalytic domain, their sequences are more similar to each other than to the other members of the subfamily. Exceptions are the three 1,3-ß-glucanases of Clo. thermocellum, T. maritima and T. neapolitana: the multiple binding modules are members of subfamily CBM4A; however, the C-terminally located binding domains TnLamA(2) (=Lam16A/CBM4-2), TmLamA(2) and LicA(4) form a clearly separated subgroup within the subfamily. The significance of this sequence divergence at the C terminus of the enzyme is not clear, but from the binding of the TnLam16A/CBM4-2 module to insoluble 1,3-ß-glucan it can be assumed that they might be responsible for the binding to the insoluble natural substrates like cell walls or insoluble aggregates of 1,3-ß-glucan.

The binding pattern of seven CBMs connected with four different hydrolytic enzymes has been determined, at least one for each subfamily (Table 4). It is obvious that these binding modules bind exactly the substrates their hydrolytic partners in the modular enzymes are able to degrade. The deep branches in the dendrogram of the CBMs may thus reflect the specificity of the catalytic domains connected with them, suggesting that they are in accordance with the sequence diversity between the domains.


View this table:
[in this window]
[in a new window]
 
Table 4. Binding specificity of the four subfamilies of CBM4

 
A similar diversity of binding substrates between subfamilies has been observed also within other CBM families e.g. in CBM2 (Tomme et al., 1995 ). Despite being different from one subfamily to another, the substrate-binding specificity might be grossly conserved within a subfamily (e.g. for 1,3-ß bonds). The clustering of the CBM binding specificity with the connected catalytic activities suggests that CBM4A modules generally might be associated with 1,3-ß-glucanases, CBM4B with endo-1,4-ß-glucanases, CBM4C with endo-1,4-ß-xylanases and -glucanases, and CBM4D with cellulases (cellobiohydrolases and cellulases with undefined specificity).

The sequence of the Lam16A binding modules was compared to the secondary structure of the homologous CfCenC/CBM4-1. This module is comprised of two five-stranded ß-sheets that fold into a jellyroll-type ß-sandwich (Johnson et al., 1996 ). It is so far the only structure of a CBM binding to soluble substrates: an oligosaccharide molecule binds into a cleft that runs across one ß-sheet face of the protein, whereas the modules binding to insoluble substrates have flat binding surfaces (Bayer et al., 1998 ). The 10 ß-sheet regions in CfCenC/CBM4-1 are generally well conserved in family CBM4 (Fig. 1). It can therefore be expected that all CBM4 members have a similar tertiary structure to CfCenC/CBM4-1. Yet the non-polar and hydrophilic amino acid residues proposed to be involved in substrate binding within the binding cleft of CfCenC/CBM4-1 are not very well conserved between the subfamilies, as could be expected from the differences in binding specificity. The calcium-binding amino acid residues in ß-sheets B2 and B3 of CfCenC(1) are involved in stabilizing the protein structure (Johnson et al., 1998 ) and are highly conserved in almost all CBMs of family 4.

In summary, the present work demonstrates that two 1,3-ß-glucan binding modules bound differently to various types of binding substrates: CBM4-1 was more specific and bound only soluble 1,3-ß-glucans; CBM4-2 was rather non-specific and bound more tightly to soluble ß-glucans, in addition it also bound to insoluble 1,3-ß-glucans. By sequence similarity the two modules were assigned to one of four newly defined subfamilies of family CBM4, members of which differ greatly in binding specificity to a variety of glycans, either soluble or insoluble, consisting of glucose or xylose monomers and connected by 1,3-ß-, 1,4-ß- or 1,6-ß-glycosidic bonds. The data presented above corroborate the hypothesis that the similarity tree reflects an evolutionary development within the CBM4 family, adapting the subfamilies to bind different substrates. A co-evolution of the non-catalytic modules with their attached catalytic domains can be assumed.


   ACKNOWLEDGEMENTS
 
This work was supported by grants from INTAS (94-1328 and YCF 98-78) and RFFI (ref. no. 00-04-48197) to I.Y.V. and G.A.V. and from Leonhard-Lorenz-Stiftung (407/98) to W.H.S. We are very grateful to W. L. Staudenbauer and B. Henrissat for useful comments and for critically reading the manuscript.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Abou Hachem, M., Nordberg Karlsson, E., Bartonek-Roxa, E., Raghothama, S., Simpson, P. J., Gilbert, H. J., Williamson, M. P. & Holst, O.(2000). Carbohydrate-binding modules from the thermostable Rhodothermus marinus xylanase: cloning, expression and binding studies. Biochem J 345, 53-60.[Medline]

Aida, K., Okada, T., Kasahara, N., Nikaidou, N., Tanaka, H. & Watanabe, T.(1995). Comparative studies of ß-1,3-glucanase A1 and B of Bacillus circulans WL-12: purifications and enzymatic properties. J Ferment Bioeng 80, 283-286.

Bacon, J. S. D, Farmer, V. C., Jones, D. & Taylor, I. F.(1969). The glucan components of the cell wall of baker’s yeast (Saccharomyces cerevisiae) considered in relation to its ultrastructure. Biochem J 114, 557-567.[Medline]

Bayer, E. A., Chanzy, H., Lamed, R. & Shoham, Y.(1998). Cellulose, cellulases and cellulosomes. Curr Opin Struct Biol 8, 548-557.[Medline]

Black, G. W., Rixon, J. E., Clarke, J. H., Hazlewood, G. P., Ferreira, L. M. A., Bolam, D. N. & Gilbert, H. J.(1996). Cellulose binding domains and linker sequences potentiate the activity of hemicellulases against complex substrates. J Biotechnol 57, 59-69.

Bolam, D. N., Ciruela, A., McQueen-Mason, S., Simpson, P., Williamson, M. P., Rixon, J. E., Boraston, A., Hazlewood, G. P. & Gilbert, H. J.(1998). Pseudomonas cellulose-binding domains mediate their effects by increasing enzyme substrate proximity. Biochem J 331, 775-781.[Medline]

Bronnenmeier, K., Kern, A., Liebl, W. & Staudenbauer, W. L.(1995). Purification of Thermotoga maritima enzymes for the degradation of cellulosic materials. Appl Environ Microbiol 61, 1399-1407.[Abstract]

Coutinho, P. M. & Henrissat, B.(1999). Carbohydrate-active enzymes: an integrated database approach. In Recent Advances in Carbohydrate Bioengineering , pp. 3-12. Edited by H. J. Gilbert, G. Davies, B. Henrissat & B. Svensson. Cambridge:Royal Society of Chemistry.

Coutinho, J. B., Gilkes, N. R., Warren, R. A. J., Kilburn, D. G. & Miller, R. C.Jr(1992). The binding of Cellulomonas fimi endoglucanase C (CenC) to cellulose and sephadex is mediated by the N-terminal repeats. Mol Microbiol 6, 1243-1252.[Medline]

Dakhova, O. N., Kurepina, N. E., Zverlov, V. V., Svetlichnyi, V. A. & Velikodvorskaya, G. A.(1993). Cloning and expression in Escherichia coli of Thermotoga neapolitana genes coding for enzymes of carbohydrate substrate degradation. Biochem Biophys Res Commun 194, 1359-1364.[Medline]

Gill, J., Rixon, J. E., Bolam, D. N., McQueen-Mason, S., Simpson, P. J., Williamson, M. P., Hazlewood, G. P. & Gilbert, H. J.(1999). The type II and X cellulose-binding domains of Pseudomonas xylanase A potentiate catalytic activity against complex substrates by a common mechanism. Biochem J 342, 473-480.[Medline]

Harada, T.(1992). The story of research into curdlan and the bacteria producing it. Trends Glycosci Glycotech 4, 309-317.

Henrissat, B., Teeri, T. T. & Warren, R. A. J.(1998). A scheme for designating enzymes that hydrolyse the polysaccharides in the cell wall of plants. FEBS Lett 425, 352-354.[Medline]

Huber, R. & Stetter, K. O.(1992). The order Thermotogales. In The Prokaryotes , pp. 3809-3815. Edited by A. Balows, H. G. Trüper, M. Dworkin, W. Harder & K.-H. Schleifer. New York:Springer.

Irwin, D., Shin, D.-H., Zhang, S., Barr, B. K., Sakon, J., Karplus, P. A. & Wilson, D. B.(1998). Roles of the catalytic domain and two cellulose binding domains of Thermomonospora fusca E4 in cellulose hydrolysis. J Bacteriol 180, 1709-1714.[Abstract/Free Full Text]

Jiang, G. & Vasanthan, T.(2000). MALDI-MS and HPLC quantification of oligosaccharides of lichenase-hydrolyzed water-soluble ß-glucan from ten barley varieties. J Agric Food Chem 48, 3305-3310.[Medline]

Johnson, P. E., Joshi, M. D., Tomme, P., Kilburn, D. G. & McIntosh, L. P.(1996). Structure of the N-terminal cellulose-binding domain of Cellulomonas fimi CenC determined by nuclear magnetic resonance spectroscopy. Biochemistry 35, 14381-14394.[Medline]

Johnson, P. E., Creagh, A. L., Brun, E., Joe, K., Tomme, P., Haynes, C. A. & McIntosh, L. P.(1998). Calcium binding by the N-terminal cellulose-binding domain from Cellulomonas fimi ß-1,4-glucanase CenC. Biochemistry 37, 12772-12781.[Medline]

Kanzawa, Y., Kurasawa, T., Kanegae, Y., Harada, A. & Harada, T.(1994). Purification and properties of a new exo-(1->3)-ß-D-glucanase from Bacillus circulans YK9 capable of hydrolyzing resistant curdlan with formation of only laminaribiose. Microbiology 140, 637-642.[Abstract]

Kim, Y.-T., Kim, E.-H., Cheong, C., Williams, D. L., Kim, C.-W. & Lim, S.-T.(2000). Structural characterization of ß-D-(1->3, 1->6)-linked glucans using NMR spectroscopy. Carbohydr Res 328, 331-341.[Medline]

Krah, M., Misselwitz, R., Politz, O., Thomsen, K. K., Welfle, H. & Borriss, R.(1998). The laminarinase from thermophilic eubacterium Rhodothermus marinus: conformation, stability and identification of active site carboxylic residues by site-directed mutagenesis. Eur J Biochem 257, 101-111.[Abstract]

Kulicke, W. M., Lettau, A. I. & Thielking, H.(1997). Correlation between immunological activity, molar mass, and molecular structure of different (1->3)-ß-D-glucans. Carbohydr Res 297, 135-143.[Medline]

Linder, M. & Teeri, T. T.(1997). The roles and function of cellulose-binding domains. J Biotechnol 57, 15-28.

Meissner, K., Wassenberg, D. & Liebl, W.(2000). The thermostabilizing domain of the modular xylanase XynA of Thermotoga maritima represents a novel type of binding domain with affinity for soluble xylan and mixed-linkage ß-1,3/ß-1,4-glucan. Mol Microbiol 36, 898-912.[Medline]

Nogi, Y. & Horikoshi, K.(1990). A thermostable alkaline ß-1,3-glucanase produced by alkalophilic Bacillus sp. AG-430. Appl Microbiol Biotechnol 32, 704-707.

Schwarz, W. H., Schimming, S. & Staudenbauer, W. L.(1988). Isolation of a Clostridium thermocellum gene encoding a thermostable ß-1,3-glucanase (laminarinase). Biotechnol Lett 10, 225-230.

Sunna, A., Gibbs, M. D. & Bergquist, P. L.(2000). The thermostabilizing domain, XynA, of Caldibacillus cellulovorans xylanase is a xylan binding domain. Biochem J 346, 583-586.[Medline]

Thompson, J. D., Higgins, D. G. & Gibson, T. J.(1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673-4680.[Abstract]

Tomme, P., Warren, R. A. J. & Gilkes, N. R.(1995). Cellulose hydrolysis by bacteria and fungi. Adv Microb Physiol 37, 1-81.[Medline]

Tomme, P., Creagh, A. L., Kilburn, D. G. & Haynes, C. A.(1996). Interaction of polysaccharides with the N-terminal cellulose-binding domain of Cellulomonas fimi CenC. I. Binding specificity and calorimetric analysis. Biochemistry 35, 13885-13894.[Medline]

Tomme, P., Boraston, A., McLean, B. & 7 other authors (1998). Characterization and affinity applications of cellulose-binding domains. J Chromatogr B715, 283–296.

Wood, T. M. & Bhat, K. M.(1988). Methods for measuring cellulase activities. Methods Enzymol 160, 87-112.

Zverlov, V. V., Volkov, I. Y., Velikodvorskaya, T. V. & Schwarz, W. H.(1997). Highly thermostable endo-1,3-ß-glucanase (laminarinase) LamA from Thermotoga neapolitana: nucleotide sequence of the gene and characterization of the recombinant gene product. Microbiology 143, 1701-1708.[Abstract]

Zverlov, V. V., Volkov, I. Y., Lunina, N. A. & Velikodvorskaya, G. A.(1999). Enzymes of thermophilic anaerobic bacteria hydrolyzing cellulose, xylan, and other beta-glucans. Mol Biol 33, 89-95.

Received 22 September 2000; revised 20 November 2000; accepted 23 November 2000.