Structural and Biochemical Analysis of Cellvibrio japonicus Xylanase 10C

Microbial degradation of the plant cell wall is the primary mechanism by which carbon is utilized in the biosphere. The hydrolysis of xylan, by endo-β-1,4-xylanases (xylanases), is one of the key reactions in this process. Although amino acid sequence variations are evident in the substrate binding cleft of “family GH10” xylanases (see afmb.cnrs-mrs.fr/CAZY/), their biochemical significance is unclear. The Cellvibrio japonicus GH10 xylanase CjXyn10C is a bi-modular enzyme comprising a GH10 catalytic module and a family 15 carbohydrate-binding module. The three-dimensional structure at 1.85 Å, presented here, shows that the sequence joining the two modules is disordered, confirming that linker sequences in modular glycoside hydrolases are highly flexible. CjXyn10C hydrolyzes xylan at a rate similar to other previously described GH10 enzymes but displays very low activity against xylooligosaccharides. The poor activity on short substrates reflects weak binding at the -2 subsite of the enzyme. Comparison of CjXyn10C with other family GH10 enzymes reveals “polymorphisms” in the substrate binding cleft including a glutamate/glycine substitution at the -2 subsite and a tyrosine insertion in the -2/-3 glycone region of the substrate binding cleft, both of which contribute to the unusual properties of the enzyme. The CjXyn10C-substrate complex shows that Tyr-340 stacks against the xylose residue located at the -3 subsite, and the properties of Y340A support the view that this tyrosine plays a pivotal role in substrate binding at this location. The generic importance of using CjXyn10C as a template in predicting the biochemical properties of GH10 xylanases is discussed. Microbial degradation of the plant cell wall is the primary mechanism by which carbon is utilized in the biosphere. The hydrolysis of xylan, by endo-β-1,4-xylanases (xylanases), is one of the key reactions in this process. Although amino acid sequence variations are evident in the substrate binding cleft of “family GH10” xylanases (see afmb.cnrs-mrs.fr/CAZY/), their biochemical significance is unclear. The Cellvibrio japonicus GH10 xylanase CjXyn10C is a bi-modular enzyme comprising a GH10 catalytic module and a family 15 carbohydrate-binding module. The three-dimensional structure at 1.85 Å, presented here, shows that the sequence joining the two modules is disordered, confirming that linker sequences in modular glycoside hydrolases are highly flexible. CjXyn10C hydrolyzes xylan at a rate similar to other previously described GH10 enzymes but displays very low activity against xylooligosaccharides. The poor activity on short substrates reflects weak binding at the -2 subsite of the enzyme. Comparison of CjXyn10C with other family GH10 enzymes reveals “polymorphisms” in the substrate binding cleft including a glutamate/glycine substitution at the -2 subsite and a tyrosine insertion in the -2/-3 glycone region of the substrate binding cleft, both of which contribute to the unusual properties of the enzyme. The CjXyn10C-substrate complex shows that Tyr-340 stacks against the xylose residue located at the -3 subsite, and the properties of Y340A support the view that this tyrosine plays a pivotal role in substrate binding at this location. The generic importance of using CjXyn10C as a template in predicting the biochemical properties of GH10 xylanases is discussed. The microbial degradation of the plant cell wall is an important biological and industrial process. The enzymatic hydrolysis of this composite structure, which comprises the most abundant source of organic carbon in the biosphere, releases photosynthetically fixed carbon that is utilized as an energy and carbon source by a range of organisms. The plant cell wall consists primarily of a complex mixture of polysaccharides of which xylan is a significant component (1Brett C.T. Waldren K. Black M. Charlewood B. Physiology and Biochemistry of Plant Cell Walls: Topics in Plant Functional Biology. 1. Chapman and Hall, London1996Google Scholar). The backbone of xylan, a β-1,4-linked xylopyranose polymer, is decorated at the O2 and/or O3 positions with arabinofuranose, acetyl, and 4-methyl-d-glucuronic acid (4-O-MeGlcA) 1The abbreviations used are: 4-O-MeGlcA, 4-O-methyl-d-glucuronic acid; CjXyn10A, C. japonicus xylanase 10A; CjXyn10C, C. japonicus xylanase 10C; CjXyn10C-m, truncated CjXyn10C consisting of the CBM15 linked to the catalytic module; CjXyn10C-GH10, catalytic module of CjXyn10C; CfXyn10A, C. fimi xylanase 10A; GH, glycoside hydrolase family; X3, xylotriose; X4, xylotetraose; X5, xylopentaose; X6, xylohexaose; PNPX2, 4-nitrophenyl-β-xylobioside; PNPX, 4-nitrophenyl-β-xyloside; PNPG2, 4-nitrophenyl-β-cellobioside; GX, glucuronoxylan; d.p., degree of polymerization; CBM, carbohydrate binding module; r.m.s., root mean square. side chains (1Brett C.T. Waldren K. Black M. Charlewood B. Physiology and Biochemistry of Plant Cell Walls: Topics in Plant Functional Biology. 1. Chapman and Hall, London1996Google Scholar). The xylan backbone is hydrolyzed by endo-β-1,4-xylanases (xylanases) that are located mainly in glycoside hydrolase family (GH) 10 and 11 (CAZY website available at afmb.cnrs-mrs.fr/CAZY/ (2Henrissat B. Biochem. J. 1991; 280: 309-316Crossref PubMed Scopus (2624) Google Scholar)). Xylanases, in common with other plant cell wall hydrolases, are generally modular enzymes in which the GH10 or GH11 catalytic modules are joined, via a linker region, to non-catalytic carbohydrate binding modules (CBMs (3Millward-Sadler S.J. Poole D.M. Henrissat B. Hazlewood G.P. Clarke J.H. Gilbert H.J. Mol. Microbiol. 1994; 11: 375-382Crossref PubMed Scopus (89) Google Scholar, 4Ferreira L.M. Durrant A.J. Hall J. Hazlewood G.P. Gilbert H.J. Biochem. J. 1990; 269: 261-264Crossref PubMed Scopus (74) Google Scholar, 5Kellett L.E. Poole D.M. Ferreira L.M. Durrant A.J. Hazlewood G.P. Gilbert H.J. Biochem. J. 1990; 272: 369-376Crossref PubMed Scopus (149) Google Scholar)), whose location varies between enzymes. GH10 xylanases have a typical (β/α)8 barrel fold and hydrolyze glycosidic bonds with net retention of anomeric configuration. The catalytic acid-base and nucleophile residues of these enzymes are located at the end of β-barrels 4 and 7, respectively (6Harris G.W. Jenkins J.A. Connerton I. Cummings N. Lo Leggio L. Scott M. Hazlewood G.P. Laurie J.I. Gilbert H.J. Pickersgill R.W. Structure. 1994; 2: 1107-1116Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar, 7Lo Leggio L. Kalogiannis S. Bhat M.K. Pickersgill R.W. Proteins. 1999; 36: 295-306Crossref PubMed Scopus (77) Google Scholar). Consistent with their endo-mode of action the substrate binding cleft of GH10 enzymes extend along the length of the proteins and can accommodate from four to seven xylose residues (8Charnock S.J. Spurway T.D. Xie H. Beylot M.H. Virden R. Warren R.A. Hazlewood G.P. Gilbert H.J. J. Biol. Chem. 1998; 273: 32187-32199Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 9Biely P. Kratky Z. Vrsanska M. Eur. J. Biochem. 1981; 119: 559-564Crossref PubMed Scopus (99) Google Scholar). Each region that accommodates a xylose moiety is known as a subsite. Subsites are given a negative or positive number depending on whether they bind the glycone or aglycone regions of the substrate, respectively. Glycosidic bond cleavage thus occurs between the -1 and +1 subsites (10Davies G.J. Wilson K.S. Henrissat B. Biochem. J. 1997; 321: 557-559Crossref PubMed Scopus (848) Google Scholar). The crystal structures of several GH10 xylanases in complex with substrate or mechanistic inhibitors have revealed highly conserved amino acids, particularly in the -1 and -2 subsites, that play a key role in substrate recognition (11Ducros V. Charnock S.J. Derewenda U. Derewenda Z.S. Dauter Z. Dupont C. Shareck F. Morosoli R. Kluepfel D. Davies G.J. J. Biol. Chem. 2000; 275: 23020-23026Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar, 12Notenboom V. Birsan C. Warren R.A. Withers S.G. Rose D.R. Biochemistry. 1998; 37: 4751-4758Crossref PubMed Scopus (88) Google Scholar, 13Notenboom V. Williams S.J. Hoos R. Withers S.G. Rose D.R. Nat. Struct. Biol. 1998; 5: 812-818Crossref PubMed Scopus (112) Google Scholar, 14White A. Tull D. Johns K. Withers S.G. Rose D.R. Nat. Struct. Biol. 1996; 3: 149-154Crossref PubMed Scopus (187) Google Scholar). Although x-ray crystallography has provided detailed insights into the structural basis for the catalytic activity displayed by glycoside hydrolases, there is a lack of information on the conformation of intact modular enzymes. It is generally believed that linker regions display a great deal of structural flexibility, thereby maximizing substrate accessibility when the enzymes are bound to the plant cell wall via CBMs (15Black G.W. Rixon J.E. Clarke J.H. Hazlewood G.P. Ferreira L.M. Bolam D.N. Gilbert H.J. J. Biotechnol. 1997; 57: 59-69Crossref PubMed Scopus (48) Google Scholar, 16Black G.W. Rixon J.E. Clarke J.H. Hazlewood G.P. Theodorou M.K. Morris P. Gilbert H.J. Biochem. J. 1996; 319: 515-520Crossref PubMed Scopus (76) Google Scholar). Only one crystal structure of a full-length, multimodular plant cell wall-degrading glycoside hydrolase, however, has been reported thus far (17Fujimoto Z. Kuno A. Kaneko S. Yoshida S. Kobayashi H. Kusakabe I. Mizuno H. J. Mol. Biol. 2000; 300: 575-585Crossref PubMed Scopus (98) Google Scholar), and this indeed revealed a highly flexible linker region. GH10 contains a large number of proteins, and the three-dimensional structural data information available for several members of this family, coupled with sequence comparisons, can be used to identify amino acid “polymorphisms” within the substrate binding cleft of this family of enzymes. Examples of these variations include the substitution of a glutamate in the -2 subsite (which makes an important hydrogen bond with the O2 of xylose) with glycine in ∼16% of GH10 enzymes. Similarly, a subset of GH10 xylanases contain a tyrosine whose hydroxyl group forms a steric clash with the C5-CH2OH moiety of glucose at the -2 subsite thus discriminating against binding of glucosides in this subsite (18Andrews S.R. Charnock S.J. Lakey J.H. Davies G.J. Claeyssens M. Nerinckx W. Underwood M. Sinnott M.L. Warren R.A. Gilbert H.J. J. Biol. Chem. 2000; 275: 23027-23033Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). The importance, in enzyme function, of the aromatic ring of this tyrosine and the -2 subsite glutamate/glycine substitution are currently unclear. To probe the structure of the linker region of modular xylanases and to explore the functional importance of the amino acid variations described above, the Cellvibrio japonicus xylanase CjXyn10C, a modular enzyme that comprises an N-terminal CBM15 appended to a GH10 catalytic domain (3Millward-Sadler S.J. Poole D.M. Henrissat B. Hazlewood G.P. Clarke J.H. Gilbert H.J. Mol. Microbiol. 1994; 11: 375-382Crossref PubMed Scopus (89) Google Scholar, 19Szabo L. Jamal S. Xie H. Charnock S.J. Bolam D.N. Gilbert H.J. Davies G.J. J. Biol. Chem. 2001; 276: 49061-49065Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar), was subjected to biochemical, structural, and mutagenic analysis. The data, including the three-dimensional structure of the intact two-domain enzyme at 1.85-Å resolution, show that the linker region of CjXyn10C is indeed highly flexible. Although the enzyme displays activity against xylan that is typical of GH10 xylanases, CjXyn10C displays unusually poor activity against xylooligosaccharides and aryl-β-cellobiosides that reflects weak substrate binding at the -2 subsite and steric clashes at the -1 subsite, respectively. In the -3 subsite, the aromatic ring of Tyr-340 is shown to play a pivotal role in substrate recognition. Dissecting the mechanistic implications of binding cleft variations will facilitate the use of a bioinformatics approach for predicting the biochemical properties of the extensive repertoire of glycoside hydrolases identified through genome sequencing programs. Bacterial Strains, Culture Conditions, and Plasmids—The Escherichia coli strains TUNER:pLysS (Novagen, Madison, WI) and XL1-Blue were used in this study. These organisms were cultured in Luria broth (LB) at 37 °C with aeration, unless otherwise stated. The plasmid vectors used were pCR-Blunt (Invitrogen, Paisley, UK), pET16b, and pET22b (Novagen). The plasmid pHX1, a recombinant of pET22b, encodes the CBM15 and GH10 catalytic module of CjXyn10C (designated CjXyn10C-m) and thus lacks the signal peptide, whereas pHX2, which was derived from pET16b, encodes the catalytic domain of the xylanase (CjXyn10C-GH10). The plasmid pNX1 encodes the catalytic domain of C. japonicus CjXyn10A (8Charnock S.J. Spurway T.D. Xie H. Beylot M.H. Virden R. Warren R.A. Hazlewood G.P. Gilbert H.J. J. Biol. Chem. 1998; 273: 32187-32199Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). Construction of pHX1 and pHX2—The region of the CjXyn10C gene (xyn10C) encoding CBM15 and the GH10 catalytic module (nucleotides 256-1818 of xyn10C) was amplified using the PCR and the primers 5′-GCTCATATGGTCGCTGCCAGCGAGG-3′ and 5′-AACTCGAGATGTGTATTAGTACATTGCG-3′, which contain NdeI and XhoI restriction sites, respectively. The PCR product was digested with NdeI and XhoI and cloned into similarly restricted pET22b to generate pHX1, which encodes CjXyn10C-m containing a C-terminal His6 tag. PCR was also used to amplify the region of xyn10C encoding only the catalytic module of CjXyn10C (nucleotides 730-1821 of xyn10C) using the primers 5′-AACATATGTATTCAGCCAATGTCGATCAC-3′ and 5′-AACTCGAGTCAATGTGTATTAGTACATTG-3′, and the amplified DNA was cloned into the NdeI and XhoI sites of pET16b to produce pHX2, which encodes the catalytic module of CjXyn10C containing an N-terminal His10 tag. Generation of Xyn10A and Xyn10C Mutants—Mutants of the CjXyn10A catalytic module, CjXyn10C-m and CjXyn10C-GH10, were generated by the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's instructions using pNX1, pHX1, and pHX2 as template DNA, respectively. The primers that were employed in the mutagenesis PCRs are shown in the Supplementary Material. The complete sequences of the DNA encoding the xylanase mutants were determined using an ABI 377 DNA sequencer employing T7 forward and reverse primers and the following custom sequencing primers, 5′-AACGGCGCGCGATGTACAAGTCG-3′ and 5′-CAGCTGGTGAAAGAAGTAGCGC-3′, for constructs encoding CjXyn10C-m, to confirm that only the desired mutations had been introduced. During the course of this study errors in the original manual sequencing of xyn10C, resulting in three amino acid changes in the encoded enzyme, were detected and a corrected version of this gene has now been submitted to GenBank™. Expression and Purification of C. japonicus GH10 Xylanases—E. coli strain TUNER:pLysS harboring an appropriate recombinant plasmid encoding CjXyn10C-m was cultured in LB supplemented with 50 μg ml-1 ampicillin (1,000 ml in 2-liter conical baffle flasks) at 37 °C and 180 rpm to mid-exponential phase (A600 nm 0.6) at which time the culture was cooled and incubated at 30 °C, before expression of the xylanase was induced by addition of isopropyl β-thiogalactopyranoside to a final concentration of 0.2 mm and incubated for a further 5 h. The cells were then harvested by centrifugation at 4,500 × g for 10 min at 4 °C and resuspended in 1/40th volume 20 mm Tris/HCl buffer, pH 8.0, containing 300 mm NaCl before being lysed by sonication and centrifuged (25,000 × g) for 15 min at 4 °C to produce cell-free extract. CjXyn10C-m was purified from cell-free extract by immobilized metal affinity chromatography as described previously (20Xie H. Bolam D.N. Nagy T. Szabo L. Cooper A. Simpson P.J. Lakey J.H. Williamson M.P. Gilbert H.J. Biochemistry. 2001; 40: 5700-5707Crossref PubMed Scopus (57) Google Scholar) using Talon™ resin (Clontech, Palo Alto, CA). The protein eluted from the matrix was dialyzed against 3 × 500 volumes of 10 mm Tris/HCl buffer, pH 8.0 (Buffer A), applied to a 3-ml DEAE-Tris-Acryl column (15 × 50 mm), and the xylanase eluted with a linear 0-500 mm NaCl gradient in Buffer A. The recovered protein was concentrated to ∼1 ml, using a VIVASPIN 10-kDa molecular mass cut-off centrifugal concentrator (Vivascience, Hannover, Germany) and then applied to a High Load™ Superdex™ 75 column (16 × 600 mm; Amersham Biosciences, Herts, UK) and eluted again in Buffer A containing 150 mm NaCl. The columns were run at 1 ml min-1 using the Bio-Rad biologic system. The purity of the protein was evaluated by SDS-PAGE. Protein concentration was determined from the calculated molar extinction coefficient of CjXyn10C-m and CjXyn10C-GH10 (prepared by the same method as for CjXyn10C-m) at 280 nm, which were 79,020 and 60,550 m-1 cm-1, respectively, with appropriately adjusted coefficients for mutant proteins. Wild type and mutant forms of CjXyn10A were prepared as described previously (21Armand S. Andrews S.R. Charnock S.J. Gilbert H.J. Biochemistry. 2001; 40: 7404-7409Crossref PubMed Scopus (33) Google Scholar). Enzyme Assays—Activity of CjXyn10C and CjXyn10A against aryl β-glycosides were determined as described previously (21Armand S. Andrews S.R. Charnock S.J. Gilbert H.J. Biochemistry. 2001; 40: 7404-7409Crossref PubMed Scopus (33) Google Scholar). Xylanase activity was performed essentially as described by Charnock et al. (22Charnock S.J. Lakey J.H. Virden R. Hughes N. Sinnott M.L. Hazlewood G.P. Pickersgill R. Gilbert H.J. J. Biol. Chem. 1997; 272: 2942-2951Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar) except that the release of reducing sugar was determined using the Somogyi-Nelson reagent (23Somogyi M. J. Biol. Chem. 1952; 195: 19-23Abstract Full Text PDF PubMed Google Scholar). Each assay was performed in triplicate. To evaluate the activity of the two xylanases against xylooligosaccharides, 0.02-460 nm enzyme (depending on the substrate and the activity of the enzyme derivative) were incubated with 2-14 μm substrate in 50 mm sodium phosphate/12 mm citrate buffer, pH 6.5, containing 1 mm calcium acetate for up to 200 min. At regular time intervals, a 0.5-ml aliquot was removed, enzyme was inactivated by boiling for 10 min, and the xylooligosaccharides in the samples were quantified by HPLC as described previously. The progress curves of oligosaccharide cleavage were used to determine the kcat/Km of the reaction using the following equation described by Matsui et al. (24Matsui I. Ishikawa K. Matsui E. Miyairi S. Fukui S. Honda K. J. Biochem. (Tokyo). 1991; 109: 566-569Crossref PubMed Scopus (74) Google Scholar), kt=ln[S0][St](Eq. 1) where k = (kcat/Km) × [enzyme], t represents time, and [S0] and [St] represent substrate concentrations at time 0 and t, respectively. The bond cleavage frequency (BCF) and kcat/Km data obtained from these experiments were used to calculate the ΔG of xylose binding at each of the subsites, following the method of Suganuma et al. (25Suganuma T. Matsuno R. Ohnishi M. Hiromi K. J. Biochem. (Tokyo). 1978; 84: 293-316Crossref PubMed Scopus (136) Google Scholar). The equation used to calculate the free energy of substrate binding was as follows. ΔG(kcal/mol)=RTlnkcat/Km(Xa)·BCF(Xa)bkcat/Km(Xa-1)·BCF(Xa-1)b/4183(Eq. 2) The kinetic parameters are as follows: kcat/Km (Xa) represents kcat/Km for a xylooligosaccharide of degree of polymerization (d.p.) a; BCF (Xa)b represents bond cleavage frequency for glycosidic bond b of a xylooligosaccharide of d.p. a; R is the universal gas constant (8.314 J mol-1 K-1); and T is temperature (Kelvin). Crystallization and Data Collection—Native and the acid/base mutant E385A of CjXyn10C-m, were crystallized using the hanging drop, vapor diffusion method. Pure proteins, as judged by SDS-PAGE, were washed into water, by repeated dilution (40 volumes of water), and concentrated in a VIVASPIN 10-kDa concentrator. Crystals grew in ∼2-4 days at 18 °C in drops containing 2 μl of 30 mg ml-1 protein, 1 μl of 0.2 m sodium iodide, and 20% (w/v) polyethylene glycol 3350 (Hampton Research, Aliso Viejo, CA). The E385A mutant was crystallized under the same conditions as the wild type enzyme except for the addition of 1 μl of 20 mm xylopentaose (Megazyme International, Bray, County Wicklow, Ireland). Crystals in the form of plates were harvested and cryogenically frozen for data collection. Data for native Xyn10C-m were collected at the European Synchrotron Radiation Facility beamline ID14-EH4 and that for the E385A complex with xylopentaose on the Daresbury Synchrotron Radiation Source on beamline pX 9.6, which both use Area Detector Systems Corporation Quantum-4 charge-coupled devices as detector. All data were processed and scaled with the HKL suite (26Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38573) Google Scholar) and all other computing used the CCP4 suite (27Collaborative Computational Project Number 4 Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19770) Google Scholar) unless otherwise stated. Structure Solution and Refinement—The native CjXyn10C-m structure was solved by molecular replacement using the program AMoRe. Data in the resolution range 20-3.0 Å were used, with an outer radius of Patterson integration of 25 Å, and the protein atoms only of the catalytic core domain of CjXyn10A (1clx.pdb) as the search model. There is one molecule in the asymmetric unit. Prior to refinement, 5% of the observations were immediately set aside for cross validation analysis (28Brünger A.T. Nature. 1992; 355: 472-475Crossref PubMed Scopus (3864) Google Scholar) and were used to monitor various refinement strategies such as the weighting of geometrical and temperature factor restraint and the insertion of solvent water during maximum-likelihood refinement using REFMAC (29Murshudov G.N. Vagin A.A. Dodson E.J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1997; 53: 240-255Crossref PubMed Scopus (13870) Google Scholar). Manual corrections of the model using the X-FIT routines of the program QUANTA (Accelrys) were interspersed with cycles of maximum-likelihood refinement. Following refinement of the catalytic core domain, it became evident that the CBM15 domain was also partially ordered “in-crystal.” Much of the CBM15 module was visible in density, and a partial model consisting of approximately four β-strands of this could easily be built into electron density and were included in refinement. Subsequently, these strands were used as a search model to solve the structure of a complex of CBM15 in isolation from its catalytic domain (19Szabo L. Jamal S. Xie H. Charnock S.J. Bolam D.N. Gilbert H.J. Davies G.J. J. Biol. Chem. 2001; 276: 49061-49065Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar). The intact CBM15 domain (1gny.pdb) could then be docked onto the initial four strands and the CBM15 domain included in refinement. The CBM15 moiety is more ordered on its outer ligand binding extremity due to crystal packing constraints and thus incorporation of the CBM15 module necessitated use of TLS (translation, libration, and screw rotation) refinement (30Winn M.D. Isupov M.N. Murshudov G.N. Acta Crystallogr. Sect. D Biol. Crystallogr. 2001; 57: 122-133Crossref PubMed Scopus (1651) Google Scholar), treating the catalytic domain (residues 243-606) and the ordered region of the CBM15 domain as discrete bodies. Defining which residues of a dynamically disordered CBM15 domain to include in the refinement was difficult, finally we included all residues from 97 to 238 for which 2Fobs - Fcalc density was above 0.66 sigma. For the native structure, density for the CBM15 domain was poor and residues 139-149, 179, 192-194, 202-207, and 213-214 could not be built into the very poor density. These regions are better ordered, although still highly mobile, in the complex structure described below. This mobility of the CBM15 domain is reflected in mean translational and vibrational components to the TLS parameters, which range from 0.28 Å2 and 2.1°2 for the complex to 0.25 Å2 and over 5°2 for the native structure. The “linker” region from residue 242 of the catalytic domain through to ∼239 of the CBM15 domain are also disordered, but at extremely low (1/3rd sigma) contour levels the density appears helical. These residues cannot, however, be modeled appropriately. It was considered inappropriate to incorporate “dummy” solvent molecules into areas where protein atoms were disordered and these areas thus display residual “difference” electron density. The refined native CjXyn10C-m structure, complete with CBM15 module, was used for refinement of the E385A complex with xylopentaose. Small changes in cell dimensions, Table I, necessitated initial rigid-body refinement in REFMAC (29Murshudov G.N. Vagin A.A. Dodson E.J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1997; 53: 240-255Crossref PubMed Scopus (13870) Google Scholar) with data from 20-3.5 Å. Following this refinement, incorporating TLS handling of correlated mobilities (30Winn M.D. Isupov M.N. Murshudov G.N. Acta Crystallogr. Sect. D Biol. Crystallogr. 2001; 57: 122-133Crossref PubMed Scopus (1651) Google Scholar), was performed as above, although in this case, the domain is considerably better ordered than in the native structure although still highly mobile. Xylooligosaccharides were incorporated toward the end of the refinement as an ordered xylotetraose on both the CBM15 domain and in the -4 to -1 subsites of the catalytic domain (in the kinetically insignificant -4 subsite the xylosyl moiety is indeed disordered). Coordinates have been deposited with the Macromolecular Structures data base. Data processing and refinement statistics are given in Table I.Table ICrystal structure and data quality statisticsXyn10C nativeXyn10C (E385A) plus X5DataWavelength0.9320Å0.8700ÅSourceESRF, ID14-EH4SRS, pX9.6Space groupP212121P212121Cell dimensionsa = 44.2 Å, b = 78.8 Å, c = 172.2 Åa = 44.6 Å, b = 82.7 Å, c = 170.7 ÅResolution range (high resolution shell)20.0−1.85 Å (1.92−1.85 Å)20−1.85 Å (1.92−1.85 Å)Rmerge0.088 (0.31)0.080 (0.37)Completeness %99 (98)98 (98)Mean I/σI16.4 (5.6)18.5 (4.8)Multiplicity5.0 (5.3)4.4 (4.7)StructureRcryst0.160.18Rfree0.200.23r.m.s bonds (Å)0.0160.017r.m.s. angles (degrees)1.41.4r.m.s chiral volumes (Å3)0.110.10PDB code1us31us2 Open table in a new tab CjXyn10C-GH10 (the catalytic module of the xylanase) was purified to electrophoretic homogeneity, and its biochemical properties were evaluated using xylooligosaccharides and both decorated and unsubstituted xylans as substrates. Enzyme activity increases as the degree of polymerization (d.p.) of the xylooligosaccharide substrate becomes larger (Table II). The difference in activity between xylohexaose and xylopentaose, however, is modest, implying that the sixth binding site, if it exists, interacts weakly with substrate. The activity of CjXyn10C-GH10 against xylooligosaccharides is ∼10-100 fold lower than other GH10 xylanases against these substrates (8Charnock S.J. Spurway T.D. Xie H. Beylot M.H. Virden R. Warren R.A. Hazlewood G.P. Gilbert H.J. J. Biol. Chem. 1998; 273: 32187-32199Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). Xylotetraose is cleaved primarily into xylose and xylotriose (Fig. 1), whereas Cellulomonas fimi Xyn10A (CfXyn10A; formerly known as Cex) and CjXyn10A both predominantly generate xylobiose from the tetrasaccharide (8Charnock S.J. Spurway T.D. Xie H. Beylot M.H. Virden R. Warren R.A. Hazlewood G.P. Gilbert H.J. J. Biol. Chem. 1998; 273: 32187-32199Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). It would appear, therefore, that the -3 subsite of CjXyn10C is kinetically much more significant than the +2 subsite, because the release of the mono- and tri-saccharide from xylotetraose requires that the substrate binds in subsites -3 through +1. To determine the actual ΔG of substrate binding at the -3, -2, and +2 subsites, the catalytic efficiency of CjXyn10C-GH10 against different xylooligosaccharides and aryl-β-xylosides was determined. The -3, -2, and +2 subsites have binding energies of 1.4, 2.4, and 2.1 kcal mol-1, respectively (Table III), in contrast to other GH10 xylanases where the -2 subsite dominates substrate-binding (8Charnock S.J. Spurway T.D. Xie H. Beylot M.H. Virden R. Warren R.A. Hazlewood G.P. Gilbert H.J. J. Biol. Chem. 1998; 273: 32187-32199Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). We conclude that the poor activity of CjXyn10C-GH10 against xylooligosaccharides is the result of a compromised -2 subsite.Table IIBiochemical properties of CjXyn10A and CjXyn10CEnzymekcat/Kmakcat/Km was determined by measuring progress curves at a substrate concentration below Km.Specific activitybSpecific activity is expressed as moles of product produced per mole of enzyme per min.X3cSubstrates used were as follows: X3, xylotriose; X4, xylotetraose; X5, xylopentaose; X6, xylohexaose; PNPX2, 4-nitrophenyl-β-d-xylobioside; PNPX, 4-nitrophenyl-β-d-xylopyranose; PNPG2, 4-nitrophenyl-β-d-cellobioside; OX, oat spelt xylan; RAX, rye arabinoxylan; WAX, wheat arabinoxylan; GX, glucuronoxylan.X4X5X6PNPX2PNPXPNPG2OXRAXWAXGXmin−1 m−1min−1CjXyn10C-mdCjXyn10C-m; a derivative of CjXyn10C comprising the CBM15 linked to the GH10 catalytic domain.2.1 × 106101056131212491563CjXyn10C-GH10eTruncated enzymes comprising only the GH10 catalytic modules.7.3 × 1028.5 × 1042.9 × 1061.0 × 1072.3 × 10634121159122314971359CjXyn10AeTruncated enzymes comprising only the GH10 catalytic modules.9.2 × 1037.3 × 1051.3 × 1071.2 × 1087.8 × 106293.2 × 1039887948331226a kcat/Km was determined by measuring progress curves at a substrate co