Crystal Structure of a Bacterial Family-Lil Cellulose-Binding Domain: a General Mechanism for Attachment to Cellulose

Crystal Structure of a Bacterial Family-Lil Cellulose-Binding Domain: a General Mechanism for Attachment to Cellulose

The EMBO Journal vol.15 no.21 pp.5739-5751, 1996 Crystal structure of a bacterial family-lIl cellulose-binding domain: a general mechanism for attachment to cellulose Jose Tormo1 2'3'4, Raphael Lamed5, problem, not only from the biotechnological point of view Arthur J.Chirinol,2, Ely Morag6, (cellulose is the most abundant, renewable source of Edward A.Bayer6, Yuval Shoham7 and organic compound on the planet), but also from the point Thomas A.Steitz1'2,8 of view of basic research, since the insoluble nature and inherent stability of crystalline cellulose constitute a 'Department of Molecular Biophysics and Biochemistry, challenge for its enzymatic hydrolysis. 8Department of Chemistry and 2Howard Hughes Medical Institute, The initial event in the cellulose degradation process is Yale University, New Haven, CT, USA, 5Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Ramat Aviv, the binding of the cellulolytic enzyme(s) or the entire Israel, 6Department of Membrane Research and Biophysics, microorganism to the cellulose substrate. This binding is The Weizmann Institute of Science, Rehovot, Israel and 7Department mediated by a separate domain, named cellulose-binding of Food Engineering and Biotechnology, Technion-Israel Institute of domain or CBD. CBDs appear to play a multiple role in Technology, Haifa, Israel cellulolysis. They often comprise a distinct domain of a 3Present address: Consejo Superior de Investigaciones Cientificas, free enzyme, linked to one or more catalytic domains (not Centro de Investigaci6n y Desarrollo, Jordi Girona 18-26, necessarily cellulases). In some cases, they occur in a E-08034 Barcelona, Spain discrete subunit, together with additional non-catalytic 4Corresponding author domains which serve to integrate the catalytic subunits into a multifunctional enzyme complex, the cellulosome The crystal structure of a family-III cellulose-binding (Lamed et al., 1983; Lamed and Bayer, 1988). When the domain (CBD) from the cellulosomal scaffoldin subunit cellulases or cellulosomes are attached to the cell surface, of Clostridium thermocellum has been determined at their CBDs mediate the binding of the cell to cellulose. 1.75 A resolution. The protein forms a nine-stranded In addition to their more obvious role as a targeting P sandwich with a jelly roll topology and binds a vehicle, it has been proposed that CBDs may mediate calcium ion. Conserved, surface-exposed residues map the non-hydrolytic disruption of cellulose fibers, thereby into two defined surfaces located on opposite sides of facilitating subsequent enzymatic degradation by the cata- the molecule. One of these faces is dominated by a lytic domains (Din et al., 1991, 1994a). planar linear strip ofaromatic and polar residues which Over 100 different CBD sequences have already been are proposed to interact with crystalline cellulose. The identified, which range in size from only 33 to over 170 other conserved residues are contained in a shallow amino acid residues. These CBDs can be grouped into groove, the function of which is currently unknown, distinctive families on the basis of amino acid sequence and which has not been observed previously in other similarities (Gilkes et al., 1991; Tomme et al., 1995). The families of CBDs. On the basis of modeling studies smallest and simplest type of CBD, comprising family I, combined with comparisons of recently determined is found only in fungal cellulases and contains between NMR structures for other CBDs, a general model for 33 to 36 residues. The three-dimensional structure of one the binding of CBDs to cellulose is presented. Although member of this family, CBH1-CBD, derived from the the proposed binding of the CBD to cellulose is essen- cellobiohydrolase I of Trichoderma reesei, has been deter- tially a surface interaction, specific types and combin- mined by NMR spectroscopy (Kraulis et al., 1989). The ations of amino acids appear to interact selectively secondary structure of CBH1-CBD is organized into a with glucose moieties positioned on three adjacent wedge-shaped irregular ,B sheet. One face of the molecule, chains of the cellulose surface. The major interaction is dominated by three conserved tyrosine side chains, forms characterized by the planar strip of aromatic residues, a hydrophobic and planar surface that has been shown to which align along one of the chains. In addition, polar be involved in cellulose binding (Linder et al., 1995; amino acid residues are proposed to anchor the CBD Reinikainen et al., 1995). molecule to two other adjacent chains of crystalline In contrast to the small CBDs observed in the fungal cellulose. cellulases, the CBDs of bacteria are substantially larger. Keywords: cellulose-binding domain (CBD)/cellulosome The structure of one of these, CexCBD, a 110-residue Clostridium thermocellum/crystal structure/scaffoldin member offamily-IT CBDs derived from a P-1 ,4-glycanase of Cellulomonas fimi, has recently been solved using NMR spectroscopy (Xu et al., 1995). CeXCBD forms an elongated, nine-stranded x barrel, and the substrate-binding Introduction site appears to include three solvent-exposed tryptophans, Cellulose, the major polysaccharide component of plant together with other hydrophilic residues, located on one cell walls, is degraded in nature by the concerted action edge of the barrel. of a number of bacterial and fungal organisms (Beguin and Family-Ill CBDs comprise -150 amino acid residues. Aubert, 1994). Cellulose degradation poses an interesting They have been identified in many different bacterial Oxford University Press 5739 J.Tormo et aL enzymes, and in the non-hydrolytic proteins CbpA (Shoseyov et al., 1992), CipA (Gemgross et al., 1993), Table I. Model refinement CipB (Poole et al., 1992) and CipC (Pages et al., 1996) Resolution range (A) 10.0-1.75 which are responsible for the structural organization of R factor (%) 19.3 the cellulosomes present in Clostridium cellulovorans Number of reflections with F>2a 26 090 Number of protein atoms 2436 (CbpA), Clostridium thermocellum (CipA and CipB Number of calcium ions 2 from strains ATCC 27405 and YS, respectively), and Number of water molecules 280 Clostridium cellulolyticum (CipC). These non-hydrolytic Average temperature factor (A2) cellulosomal structural proteins, named scaffoldins (Bayer main chain atoms 27.0 et al., 1994), consist of a single large polypeptide chain side chain atoms 29.6 calcium atoms 22.8 of similar size (-1800 amino acid residues). Besides a R.m.s. deviations B-factors between bonded atoms (A2) 2.4 discrete number of domains which interact with different R.m.s. deviations from ideal values catalytic subunits to form the cohesive cellulosome struc- bond lengths (A) 0.010 ture, the scaffoldins characterized so far contain a single bond angles (0) 1.45 dihedral angles (0) 28.67 CBD. improper torsion angles (0) 1.40 In this communication, we report the high-resolution R.m.s. deviations between subunits crystal structure of the CBD from the cellulosomal scaf- main-chain atoms (A) 0.23 foldin subunit of C.thermocellum. The structure of this all atoms (A) 0.49 family-III CBD is compared and contrasted with other B-factors equivalent atoms (A2) 1.83 known CBD structures, highlighting important novel features. We propose its mode of binding and interaction with the cellulose substrate, based on available structural and all atoms, respectively. The major differences between and mutagenesis data for this and related CBDs. the two copies are localized in flexible loops connecting P strands (loops 5-6 and 8-9). These solvent-exposed Results regions form the rims of the concave groove on one sheet, and are in different crystal environments. Crystal structure determination Throughout the discussion, amino acids in Cip-CBD The CBD of the scaffoldin subunit Cip (Cip-CBD) of the shall be numbered starting at the first residue which shows cellulosome from C.thermocellum (comprising residues electron density as Asnl, which corresponds to residue 361 to 527) was expressed in Escherichia coli, purified 368 in CipA from C.thermocellum strain ATCC 27405 and crystallized as previously described (Lamed et al., (Gerngross et al., 1993). The last residue observed in the 1994; Morag et al., 1995). The crystals were grown by electron density maps is Prol55 (residue 522 in CipA). vapor diffusion using PEG as a precipitant. They belong This residue is well conserved in family-III CBDs, and to the monoclinic space group C2, and contain two constitutes the ultimate or penultimate amino acid in those molecules in the asymmetric unit. The crystal structure of proteins where the CBD is located at the C-terminus. Cip-CBD was solved by, conventional multiple iso- morphous replacement including anomalous scattering Description of the overall structure of Cip-CBD (MIRAS) techniques, using two heavy atom derivatives, The crystal structure shows that 155 amino acid residues with data extending to 2.5A resolution (see Materials and of Cip-CBD fold into a single, compact domain that has methods). The MIRAS phases, although of good quality, an overall prismatic shape with approximate dimensions were further improved by solvent flattening, non-crystallo- of 30 Ax3O Ax45 A. Cip-CBD belongs to the all-f graphic symmetry averaging, and histogram matching. family of proteins and is arranged in two antiparallel 3 The combination of these procedures provided a high- sheets that stack face-to-face to form a 3 sandwich with quality electron density map which was readily interpret- jelly roll topology (Figures 1 and 2). Except for the nine able. The final atomic model has been refined to an R major f strands arranged in the two sheets, the protein factor of 0.193 using 2a data extending from 10.0 to 1.75 A consists mostly ofloops connecting the secondary structure resolution (Table I). Deviations from ideal stereochemistry elements. However, the connections between strands 1-2, and the distribution of conformational angles about the 7-8, on one side of the molecule, and 2-3, 6-7 on the expected values are within the ranges expected for well- other side, form very short antiparallel 0 strands.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    13 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us