The family 11 Carbohydrate-Binding Module of Clostridium thermocellum Lic26A-Cel5E accommodates ββββ -1,4 and ββββ -1,3-1,4-mixed linked glucans at a single binding site

Modular glycoside hydrolases that attack recalcitrant polymers generally contain noncatalytic carbohydrate-binding modules (CBMs), which play a critical role in the action of these enzymes by localizing the appended catalytic domains onto the surface of insoluble polysaccharide substrates. Type B CBMs, which recognize single polysaccharide chains, display ligand specificities that are consistent with the substrates hydrolyzed by the associated catalytic domains. In enzymes that contain multiple catalytic domains with distinct substrate specificities, it is unclear how these different activities influence the evolution of the ligand recognition profile of the appended CBM. To address this issue, we have characterized the properties of a family 11 CBM (CtCBM11) in Clostridium thermocellum Lic26A-Cel5E, an enzyme that contains GH5 and GH26 catalytic domains that display beta-1,4- and beta-1,3-1,4-mixed linked endoglucanase activity, respectively. Here we show that CtCBM11 binds to both beta-1,4- and beta-1,3-1,4-mixed linked glucans, displaying K(a) values of 1.9 x 10(5), 4.4 x 10(4), and 2 x 10(3) m(-1) for Glc-beta1,4-Glc-beta1,4-Glc-beta1,3-Glc, Glc-beta1,4-Glc-beta1,4-Glc-beta1,4-Glc, and Glc-beta1,3-Glc-beta1,4-Glc-beta1,3-Glc, respectively, demonstrating that CBMs can display a preference for mixed linked glucans. To determine whether these ligands are accommodated in the same or diverse sites in CtCBM11, the crystal structure of the protein was solved to a resolution of 1.98 A. The protein displays a beta-sandwich with a concave side that forms a potential binding cleft. Site-directed mutagenesis revealed that Tyr(22), Tyr(53), and Tyr(129), located in the putative binding cleft, play a central role in the recognition of all the ligands recognized by the protein. We propose, therefore, that CtCBM11 contains a single ligand-binding site that displays affinity for both beta-1,4- and beta-1,3-1,4-mixed linked glucans.


INTRODUCTION
The major plant cell wall polysaccharides, cellulose and hemicellulose, are hydrolysed into soluble sugars by a consortium of microbial enzymes. In nature, these enzyme systems play a major role in recycling plant cell wall fixed carbon and are therefore of considerable biological and biotechnological importance. In general, glycoside hydrolases (GH) that catalyse plant cell wall degradation contain non-catalytic carbohydrate binding modules (CBMs) that interact with polysaccharides (1). Based on primary structure similarity, CBMs have been grouped into 39 different families ((2) http://afmb.cnrs-mrs.fr/CAZY). By mediating an intimate and prolonged association of the enzyme with its target substrate, CBMs enhance the activity of the catalytic module against insoluble polysaccharides (3,4).
Therefore, CBMs play a major role in potentiating the capacity of cellulases and hemicellulases to degrade plant cell wall polysaccharides.
Structural studies have revealed that the topology of the ligand-binding site of CBMs varies.
CBMs that interact with the flat surfaces of crystalline polysaccharides contain a planar hydrophobic carbohydrate binding site and are referred to as Type A CBMs (5)(6)(7)(8). In contrast, CBMs that bind to single polysaccharide chains accommodate these ligands in extended clefts and are defined as Type B CBMs (9)(10)(11)(12). Type B CBMs have been described that bind to a diversity of ligands, with some modules displaying plasticity in their capacity to accommodate heterogeneity in the sugar backbone both in terms of the identity of the saccharides and the nature of the linkage. Thus, polar residues in CBM29 are able to hydrogen bond to the axial O2 in mannose and the equatorial O2 in glucose enabling the protein to bind cellulose, glucomannan and mannan (13). Some CBMs from families 4 and 6 display affinity for the β-1,3-1,4mixed linked glucans (barley β-glucan and lichenan) and β-1,4 glucans such as amorphous cellulose and cellohexaose (14,15). It is currently unclear by guest on March 24, 2020 http://www.jbc.org/ Downloaded from whether these proteins interact exclusively with the β-1,4 regions of the mixed linked polysaccharide or display specificity for sequences of glucose units that are linked by a mixture of β-1,3 and β-1,4 glycosidic bonds. In general the ligand specificity of Type B CBMs reflects the substrate hydrolysed by the associated catalytic modules (1). In enzymes that contain multiple catalytic domains with distinct substrate specificities, however, it is unclear how these different activities influence the evolution of the ligand recognition profile of the appended CBM.
To investigate whether CBMs can display enhanced affinity for mixed linked glucans and to interrogate the evolutionary pressure exerted on the ligand specificity of CBMs in enzymes that contain distinct catalytic modules, we have studied the structure and function of the family 11 CBM (CtCBM11) in Clostridium thermocellum Lic26A-Cel5E (16). This enzyme contains GH5 and GH26 catalytic domains that display β-1,4 and β-1,3-1,4-mixed-linked endoglucanase activity, respectively. Here we show that CtCBM11 does indeed display a preference for specific β-1,3-1,4-mixed linked glucans although the protein is able to bind to β-1,4 glucose polymers. The 3D structure of CtCBM11 in harness with mutagenesis studies reveals that the protein contains a single ligand binding cleft that can accommodate both β-1,3-1,4-and β-1,4linked glucans.

Protein Expression and Purification
To express CtCBM11 in Escherichia coli, the region of the Lic26A-Cel5A gene (lic26A-cel5A) encoding the internal family 11  hours. Cells were collected by centrifugation and the cell pellet was resuspended in a 50 mM sodium Hepes buffer, pH 7.5, containing 1 M NaCl and 10 mM imidazole. For biochemical assays, recombinant CtCBM11 was purified by immobilized metal ion affinity chromatography as described previously (17).

Source of sugars used
All soluble polysaccharides were purchased from Megazyme International (Bray County Wicklow, Ireland), except oat spelt xylan and hydroxyethylcellulose, which were obtained

Affinity gel electrophoresis
The affinity of CtCBM11 for a range of soluble polysaccharides was determined by affinity gel electrophoresis (AGE). The method was essentially as described by Tomme et al. (20) using the polysaccharide ligands at a concentration of 0.1 % (w/v). Electrophoresis was carried out for 4 h in native polyacrylamide gels containing 10% (w/v) acrylamide. The nonbinding negative control, was bovine serum albumin. Quantitative assessment of binding was carried out as described previously (21). concentration of CtCBM11 binding sites present in the polysaccharide ligands was determined as described previously (11). Integrated heat effects, after correction for heats of dilution, were analyzed by nonlinear regression using a single-site binding model (Microcal Titrations were carried out in triplicate for most ligands, and the errors are the S.D. of the mean of these replicates.

Site-directed mutagenesis
Mutants of CtCBM11 were generated using the PCR-based QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. The primers used to generate these mutants were as follows: TGGGGTTCAGCCTCCGGTGAAGGTGC and TGCACCTTCACCGGAGGCTGAACCCCA, Y22A;

Production and purification of seleno-L-methionine containing CtCBM11
The methionine auxotroph E. coli B834 (DE3), transformed with pAG1, was cultured at 37ºC in one litre of culture medium as described by Charnock et al. (9). Expression of CtCBM11 was induced by the addition of 1 mM IPTG when cells were at mid-exponential growth phase. Cells were incubated at 37ºC for a further 16 h, after which time were collected and the recombinant protein purified by affinity chromatography as described above. For crystallization trials, gel filtration was included as a further purification step. The enzyme was buffer exchanged, using a PD-10 Sephadex G-25M gel filtration columns (Amersham Biosciences), into 50 mM Hepes buffer, pH 7.5, containing 200 mM NaCl (Buffer A), concentrated to 20 mg/ml with Amicon 10 kDa molecular-weight centrifugation membranes, and subjected to gel filtration using a HiLoad 16/60 Superdex 75 column (Amersham Biosciences) with protein eluted at 1 ml/min in Buffer A. Purified enzyme was concentrated, as described before, washed three times with 5 mM DTT using the same centrifugal membranes, and the final protein concentration was adjusted to 50 mg/ml.

Crystallization and data collection
Crystals of seleno-L-methionine-containing protein were grown by vapor-phase diffusion  Table 1.

Phasing, model building and refinement
Four selenium sites were located and refined with SOLVE (24), using the peak (0.9810 Å) and the edge (0.9813 Å) data sets, which showed a high correlation of anomalous differences. and Ala 2 of the N-terminus and 3 histidine residues of the C-terminus, part of the His tag.

Ligand specificity of CtCBM11
The family 11 CBM of endoglucanase F of Fibrobacter succinogenes S85 was previously shown to bind Avicel (29). It was based on this observation that members of this family were classified as CBMs. To characterize the ligand binding specificity of CtCBM11, the protein showed that CtCBM11 displays highest affinity for β-1,3-1,4-glucans, while exhibiting significantly weaker binding to hydroxyethyl cellulose, glucomannan and oat spelt xylan (see Table 3). CtCBM11 did not associate with arabinan, galactomannan, laminarin, rhamnogalacturan, glucuronoxylan or rye-arabinoxylan.
To explore the ligand specificity of CtCBM11 in more detail, the binding of the protein to polysaccharides and oligosaccharides was evaluated using ITC. Example titrations, displayed in Figure 1, reveal sigmoidal titration curves for β-1,3-1,4 and β-1,4-glucooligosaccharides as well as the mixed linked polysaccharides barley β-glucan and lichenan, enabling the K a , ∆H and stoichiometry of binding to be accurately determined for these ligands ( Table 2). The data show that CtCBM11 displays similar affinity for cellohexaose, cellopentaose and cellotetraose. Although the protein retains significant affinity for cellotriose (K a 1.6 × 10 4 ), binding to cellobiose is too weak to accurately quantify by ITC (~1.3 × 10 3 ). These data suggest that CtCBM11 contains at least 4 sugar binding sites, which is consistent with the ligands accommodated by other Type B CBMs. Interestingly, the protein displays approximately 4-fold higher affinity for the two β-1,3-1,4-mixed linked glucan polymers indicating that the CBM displays a preference for a β-1,3-linked glucose in at least one sugar binding site. To investigate this bond preference in more detail the capacity of CtCBM11 to bind a range of oligosaccharides was assessed. The protein displayed no significant affinity for laminohexaose, a β-1,3-linked oligosaccharide of glucose, indicating that not all the sugar binding sites in the CBM can accommodate β-1,3-linked glucose residues (data not shown).
Glc-β1,3-Glc approximately four times more tightly than cellotetraose, supporting the view that the protein displays a preference for a β-1,3-linked glucose in at least one subsite. A second mixed linkage glucotetra-oligosaccharide, Glc-β1,3-Glc-β1,4-Glc-β1,3-Glc, bound only weakly to CtCBM11 (K a ~2 × 10 3 ; affinity too low to accurately quantify by ITC) suggesting that the protein may only be able to accommodate a single β-1,3-linked glucose. The

Crystal structure of CtCBM11 and its similarity with other CBMs
The structure of the CtCBM11 was solved using MAD methods with a seleno-methionine derivative crystal. Experimental phases were obtained to a resolution of 1.98 Å, which generated electron density maps of sufficiently high quality for automatic interpretation.
Statistics of the diffraction data and the final model are presented in Table 1
The      Fibrobacter succinogenes (FsCBM11). Identity to CtCBM11 is indicated with grey boxes.
Residue numbers refer to the sequence of CtCBM11, which was purified and crystallized.
Residues in CtCBM11 that have side chains inside the putative binding cleft are marked with dark grey boxes. The depicted secondary structure corresponds to CtCBM11. The sequence alignment was calculated with program CLUSTALW (43) and the picture was produced with program ALSCRIPT (44).