Anaerostipes caccae gen. nov., sp. nov., a new saccharolytic, acetate-utilising, butyrate-producing bacterium from human faeces.

Two strains of a previously undescribed Eubacterium-like bacterium were isolated from human faeces. The strains are Gram-variable, obligately anaerobic, catalase negative, asporogenous rod-shaped cells which produced acetate, butyrate and lactate as the end products of glucose metabolism. The two isolates displayed 99.9% 16S rRNA gene sequence similarity to each other and treeing analysis demonstrated the faecal isolates are far removed from Eubacterium sensu stricto and that they represent a new subline within the Clostridium coccoides group of organisms. Based on phenotypic and phylogenetic criteria, it is proposed that the two strains from faeces be classified as a new genus and species, Anaerostipes caccae. The type strain of Anaerostipes caccae is NCIMB 13811T (= DSM 14662T).

Binding site analysis of cellulose binding domain CBD(N1) from endoglucanse C of Cellulomonas fimi by site-directed mutagenesis.

Endoglucanase C (CenC), a beta1,4 glucanase from the soil bacterium Cellulomonas fimi, binds to amorphous cellulose via two homologous cellulose binding domains, termed CBD(N1) and CBD(N2). In this work, the contributions of 10 amino acids within the binding cleft of CBD(N1) were evaluated by single site-directed mutations to alanine residues. Each isolated domain containing a single mutation was analyzed for binding to an insoluble amorphous preparation of cellulose, phosphoric acid swollen Avicel (PASA), and to a soluble glucopyranoside polymer, barley beta-glucan. The effect of any given mutation on CBD binding was similar for both substrates, suggesting that the mechanism of binding to soluble and insoluble substrates is the same. Tyrosines 19 and 85 were essential for tight binding by CBD(N1) as their replacement by alanine results in affinity decrements of approximately 100-fold on PASA, barley beta-glucan, and soluble cellooligosaccharides. The tertiary structures of unbound Y19A and Y85A were assessed by heteronuclear single quantum coherence (HSQC) spectroscopy. These studies indicated that the structures of both mutants were perturbed but that all perturbations are very near to the site of mutation.

Synthesis and biological evaluation of 1,1-difluoro-2-(tetrahydro-3-furanyl)ethylphosphonic acids possessing a N9-purinylmethyl functional group at the ring. a new class of inhibitors for purine nucleoside phosphorylases.

1,1-Difluoro-2-(tetrahydro-3-furanyl)ethylphosphonic acids cis-3 and trans-3 possessing a N9-purinylmethyl functionality at the ring were synthesized and tested as "multi-substrate analogue" inhibitors for purine nucleoside phosphorylases. Radical cyclization of allyic alpha,alpha-difluorophosphonate (E)-7 was applied to construct the alpha,alpha-difluorophosphonate-functionalized tetrahydrofuranyl moiety. The IC50 values of cis-3 and trans-3 for human erythrocyte PNP-catalyzed phosphorylation of inosine were determined to be 88 and 320 nM, respectively. The stereochemistry of the inhibitors was found to affect significantly the inhibitory potency.

Sequence of subunit a of the Na(+)-translocating F1F0-ATPase of Acetobacterium woodii: proposal for residues involved in Na+ binding.

Na+ transport through the F0 domain of Na(+)-F1F0-ATPases involves the combined action of subunits c and a but the residues involved in Na+ liganding in subunit a are unknown. As a first step towards the identification of these residues, we have cloned and sequenced the gene encoding subunit a of the Na(+)-F1F0-ATPase of Acetobacterium woodii. This is the second sequence available now for this subunit from Na(+)-F1F0-ATPases. A comparison of subunit a from Na(+)-F1F0-ATPases with those from H(+)-translocating enzymes unraveled structural similarity in a C-terminal segment including the ultimate and penultimate transmembrane helix. Seven residues are conserved in this region and, therefore, likely to be involved in Na+ liganding.

Mannan-degrading enzymes from Cellulomonas fimi.

The genes man26a and man2A from Cellulomonas fimi encode mannanase 26A (Man26A) and beta-mannosidase 2A (Man2A), respectively. Mature Man26A is a secreted, modular protein of 951 amino acids, comprising a catalytic module in family 26 of glycosyl hydrolases, an S-layer homology module, and two modules of unknown function. Exposure of Man26A produced by Escherichia coli to C. fimi protease generates active fragments of the enzyme that correspond to polypeptides with mannanase activity produced by C. fimi during growth on mannans, indicating that it may be the only mannanase produced by the organism. A significant fraction of the Man26A produced by C. fimi remains cell associated. Man2A is an intracellular enzyme comprising a catalytic module in a subfamily of family 2 of the glycosyl hydrolases that at present contains only mammalian beta-mannosidases.

Sugar ring distortion in the glycosyl-enzyme intermediate of a family G/11 xylanase.

The 1.8 A resolution structure of the glycosyl-enzyme intermediate formed on the retaining beta-1,4-xylanase from Bacillus circulans has been determined using X-ray crystallographic techniques. The 2-fluoro-xylose residue bound in the -1 subsite adopts a 2,5B (boat) conformation, allowing atoms C5, O5, C1, and C2 of the sugar to achieve coplanarity as required at the oxocarbenium ion-like transition states of the double-displacement catalytic mechanism. Comparison of this structure to that of a mutant of this same enzyme noncovalently complexed with xylotetraose [Wakarchuk et al. (1994) Protein Sci. 3, 467-475] reveals a number of differences beyond the distortion of the sugar moiety. Most notably, a bifurcated hydrogen bond interaction is formed in the glycosyl-enzyme intermediate involving Heta of Tyr69, the endocyclic oxygen (O5) of the xylose residue in the -1 subsite, and Oepsilon2 of the catalytic nucleophile, Glu78. To gain additional understanding of the role of Tyr69 at the active site of this enzyme, we also determined the 1.5 A resolution structure of the catalytically inactive Tyr69Phe mutant. Interestingly, no significant structural perturbation due to the loss of the phenolic group is observed. These results suggest that the interactions involving the phenolic group of Tyr69, O5 of the proximal saccharide, and Glu78 Oepsilon2 are important for the catalytic mechanism of this enzyme, and it is proposed that, through charge redistribution, these interactions serve to stabilize the oxocarbenium-like ion of the transition state. Studies of the covalent glycosyl-enzyme intermediate of this xylanase also provide insight into specificity, as contacts with C5 of the xylose moiety exclude sugars with hydroxymethyl substituents, and the mechanism of catalysis, including aspects of stereoelectronic theory as applied to glycoside hydrolysis.

The cellulose-binding domains from Cellulomonas fimi beta-1, 4-glucanase CenC bind nitroxide spin-labeled cellooligosaccharides in multiple orientations.

The N-terminal cellulose-binding domains CBDN1 and CBDN2 from Cellulomonas fimi cellulase CenC each adopt a jelly-roll beta-sandwich structure with a cleft into which amorphous cellulose and soluble cellooligosaccharides bind. To determine the orientation of the sugar chain within these binding clefts, the association of TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl-4-yl) spin-labeled derivatives of cellotriose and cellotetraose with isolated CBDN1 and CBDN2 was studied using heteronuclear 1H-15N NMR spectroscopy. Quantitative binding measurements indicate that the TEMPO moiety does not significantly perturb the affinity of the cellooligo-saccharide derivatives for the CBDs. The paramagnetic enhancements of the amide 1HN longitudinal (DeltaR1) and transverse (DeltaR2) relaxation rates were measured by comparing the effects of TEMPO-cellotetraose in its nitroxide (oxidized) and hydroxylamine (reduced) forms on the two CBDs. The bound spin-label affects most significantly the relaxation rates of amides located at both ends of the sugar-binding cleft of each CBD. Similar results are observed with TEMPO-cellotriose bound to CBDN1. This demonstrates that the TEMPO-labeled cellooligosaccharides, and by inference strands of amorphous cellulose, can associate with CBDN1 and CBDN2 in either orientation across their beta-sheet binding clefts. The ratio of the association constants for binding in each of these two orientations is estimated to be within a factor of five to tenfold. This finding is consistent with the approximate symmetry of the hydrogen-bonding groups on both the cellooligosaccharides and the residues forming the binding clefts of the CenC CBDs.

Cloning of a novel prolidase gene from Aureobacterium esteraromaticum.

The prolidase gene from Aureobacterium esteraromaticum was cloned and expressed in Escherichia coli. The cloned enzyme had the same enzymatic properties as the wild-type enzyme. Kinetic analysis of the enzyme indicated that the best substrate was Pro-Hyp, which was not hydrolyzed by other prolidases. Interestingly, there was no homology between the deduced amino acid sequence of A. esteraromaticum prolidase and those of the other sources such as human E. coli and Lactobacillus. However, homology was seen with the yeast hypothetical protein YJL213w, the function of which is unknown. These findings indicate that the A. esteraromaticum prolidase is a novel enzyme different from other prolidases reported to date.

The topology of the substrate binding clefts of glycosyl hydrolase family 10 xylanases are not conserved.

The crystal structures of family 10 xylanases indicate that the distal regions of their active sites are quite different, suggesting that the topology of the substrate binding clefts of these enzymes may vary. To test this hypothesis, we have investigated the rate and pattern of xylooligosaccharide cleavage by the family 10 enzymes, Pseudomonas fluorescens subsp. cellulosa xylanase A (XYLA) and Cellulomonas fimi exoglucanase, Cex. The data showed that Cex contained three glycone and two aglycone binding sites, while XYLA had three glycone and four aglycone binding sites, supporting the view that the topologies of substrate binding clefts in family 10 glycanases are not highly conserved. The importance of residues in the substrate binding cleft of XYLA in catalysis and ligand binding were evaluated using site-directed mutagenesis. In addition to providing insight into the function of residues in the glycone region of the active site, the data showed that the aromatic residues Phe-181, Tyr-255, and Tyr-220 play important roles in binding xylose moieties, via hydrophobic interactions, at subsites +1, +3, and +4, respectively. Interestingly, the F181A mutation caused a much larger reduction in the activity of the enzyme against xylooligosaccharides compared with xylan. These data, in conjunction with a previous study (Charnock, S. J., Lakey, J. H., Virden, R., Hughes, N., Sinnott, M. L., Hazlewood, G. P., Pickersgill, R., and Gilbert, H. J. (1997) J. Biol. Chem. 272, 2942-2951), suggest that the binding of xylooligosaccharides at the -2 and +1 subsites ensures that the substrates occupy the -1 and +1 subsites and thus preferentially form productive complexes with the enzyme. Loss of ligand binding at either subsite results in small substrates forming nonproductive complexes with XYLA by binding to distal regions of the substrate binding cleft.

Crystallization and preliminary X-ray studies of purine nucleoside phosphorylase from Cellulomonas sp.

The commercially available enzyme purine nucleoside phosphorylase (PNP) from Cellulomonas sp. was purified by ion--exchange chromatography, partially sequenced and crystallized in two different crystal forms using the hanging-drop vapour-diffusion technique. Crystal form A grows as polyeders and/or cubes in the cubic space group P4232 with unit-cell dimension a = 162.5 A. Crystal form B appears as thick plates in the space group P212121 with unit-cell dimensions a = 63.2, b = 108.3 and c = 117.4 A. Both crystal forms contain three monomers (one trimer) in the asymmetric unit.

Calcium binding by the N-terminal cellulose-binding domain from Cellulomonas fimi beta-1,4-glucanase CenC.

The interaction of the N-terminal cellulose-binding domain, CBDN1, from Cellulomonas fimi beta-1,4-glucanase CenC with calcium was investigated using NMR spectroscopy and calorimetry. CBDN1 binds a single calcium ion with an equilibrium association constant of approximately 10(5) M-1 at 35 degreesC and pH 6.0. Binding is exothermic (-42 +/- 2 kJ mol-1) under these conditions and is accompanied by a small negative change in heat capacity (DeltaCp = -0.41 +/- 0.16 kJ mol-1 K-1). From an NMR line shape analysis, the rate constants for calcium association and dissociation were found to be (5 +/- 2) x 10(7) s-1 M-1 and (4.5 +/- 0.6) x 10(2) s-1, respectively. The rapid association kinetics indicate that the calcium-binding site on CBDN1 is accessible and, to the first approximation, preformed. Based on patterns of chemical shift perturbations, and structural comparisons with the Bacillus sp. 1, 3-1,4-beta-glucanases, the backbone carbonyl oxygens of Thr8, Gly30, and Asp142 and a side chain carboxyl oxygen of Asp142 are postulated to form the calcium-binding site of CBDN1. Consistent with the calcium-independent affinity of CBDN1 for cellopentaose, this exposed site is located on the face of CBDN1 opposite to that forming the oligosaccharide-binding cleft. The midpoint denaturation temperature of CBDN1 is increased by approximately 8 degreesC at pH 6.0 in the presence of saturating amounts of calcium, confirming that metal ion binding is thermodynamically linked to native-state stability.

Cloning and expression of beta-glucosidase genes in Escherichia coli and Saccharomyces cerevisiae using shuttle vector pYES 2.0.

Genes for beta-glucosidase (Bgl) isolated from a genomic library of the cellulolytic bacterium, Cellulomonas biazotea, were cloned in pUC18 in its SacI cloning site and transformed to E. coli. Ten putative recombinants showed blackening zones on esculin plates, yellow zones on pNPG plates, in liquid culture and on native polyacrylamide gel electrophoresis activity gels. They fell into three distinct groups. Three representative E. coli clones carried recombinant plasmids designated pRM54, pRM1 and pRM17. The genes were located on 5.6-, 3.7- and 1.84-kb fragments, respectively. Their location was obtained by deletion analysis which revealed that 5.5, 3.2, and 1.8 kb fragments were essential to code for BglA, BglB, and BglC, respectively, and conferred intracellular production of beta-glucosidase on E. coli. Expression of the bgl genes resulted in overproduction of beta-glucosidase in the three clones. Secretion occurred into the periplasmic fractions. Three inserts carrying bgl genes from the representative recombinant E. coli were isolated with SacI, ligated in the shuttle vector pYES 2.0 in its SacI site and transformed to E. coli and S. cerevisiae. The recombinant plasmids were redesignated pRPG1, pRPG2 and pRPG3 coding for BglA1, BglB1 and BglC1. The cloned genes conferred extracellular production of beta-glucosidase on S. cerevisiae and enabled it to grow on cellobiose and salicin. The gall promoter of shuttle vector pYES 2.0 enabled the organisms to produce twice more beta-glucosidase than that supported by the lacZ-promoter of pUC18 plasmid in E. coli. The cloned gene can be used as a selection marker for introducing recombinant plasmids in wild strains of S. cerevisiae. The enzyme produced by bgl+ yeast and E. coli recombinants resembles that of the donor with respect to temperature and pH requirement for maximum activity. Other enzyme properties of the beta-glucosidases from S. cerevisiae were substantially the same as those from C. biazotea.

Sed1p is a major cell wall protein of Saccharomyces cerevisiae in the stationary phase and is involved in lytic enzyme resistance.

A 260-kDa structural cell wall protein was purified from sodium dodecyl sulfate-treated cell walls of Saccharomyces cerevisiae by incubation with Rarobacter faecitabidus protease I, which is a yeast-lytic enzyme. Amino acid sequence analysis revealed that this protein is the product of the SED1 gene. SED1 was formerly identified as a multicopy suppressor of erd2, which encodes a protein involved in retrieval of luminal endoplasmic reticulum proteins from the secretory pathway. Sed1p is very rich in threonine and serine and, like other structural cell wall proteins, contains a putative signal sequence for the addition of a glycosylphosphatidylinositol anchor. However, the fact that Sed1p, unlike other cell wall proteins, has six cysteines and seven putative N-glycosylation sites suggests that Sed1p belongs to a new family of cell wall proteins. Epitope-tagged Sed1p was detected in a beta-1,3-glucanase extract of cell walls by immunoblot analysis, suggesting that Sed1p is a glucanase-extractable cell wall protein. The expression of Sed1p mRNa increased in the stationary phase and was accompanied by an increase in the Sed1p content of cell walls. Disruption of SED1 had no effect on exponentially growing cells but made stationary-phase cells sensitive to Zymolyase. These results indicate that Sed1p is a major structural cell wall protein in stationary-phase cells and is required for lytic enzyme resistance.

Exploring the cellulose/xylan specificity of the beta-1,4-glycanase cex from Cellulomonas fimi through crystallography and mutation.

The retaining beta-1,4-glycanase Cex from Cellulomonas fimi, a family 10 glycosyl hydrolase, hydrolyzes xylan 40-fold more efficiently than cellulose. To gain insight into the nature of its preference for xylan, we determined the crystal structure of the Cex catalytic domain (Cex-cd) trapped as its covalent 2-deoxy-2-fluoroxylobiosyl-enzyme intermediate to 1.9 A resolution. Together with the crystal structure of unliganded Cex-cd [White, A., et al. (1994) Biochemistry 33, 12546-12552] and the previously determined crystal structure of the covalent 2-deoxy-2-fluorocellobiosyl-Cex-cd intermediate [White, A., et al. (1996) Nat. Struct. Biol. 3, 149-154], this structure provides a convincing rationale for the observed substrate specificity in Cex. Two active site residues, Gln87 and Trp281, are found to sterically hinder the binding of glucosides and must rearrange to accommodate these substrates. Such rearrangements are not necessary for the binding of xylobiosides. The importance of this observation was tested by examining the catalytic behavior of the enzyme with Gln87 mutated to Met. This mutation had no measurable effect on substrate affinity or turnover number relative to the wild type enzyme, indicating that the Met side chain could accommodate the glucoside moiety as effectively as the wild type Gln residue. Subsequent mutagenesis studies will address the role of entropic versus enthalpic contributions to binding by introducing side chains that might be more rigid in the unliganded enzyme.

Synthesis of 1,1-difluoro-5-(1H-9-purinyl)-2-pentenylphosphonic acids and the related methano analogues. Remarkable effect of the nucleobases and the cyclopropane rings on inhibitory activity toward purine nucleoside phosphorylase.

A series of 1,1-difluoro-5-(1H-9-purinyl)-2-pentenylphosphonic acids, (E)-2a,b and (Z)-2a,b, as well as the related methano analogues (+/-)-3a,b and (+/-)-4a,b were prepared for evaluation of their PNP inhibitory activities. The cyclopopane ring and the hypoxanthine residue were found to increase the profile of inhibitory activity. The IC50 and Ki values of difluoro¿(1R*,2S*)-2-[2-(6-oxo-6,9-dihydro-1H-9-purinyl)ethyl]cycl opropyl¿methylphosphonic acid (+/-)-3b toward PNP purified from Cellulomonas sp. were determined to be 70 nM and 8.8 nM, respectively.

The role of oil in macromolecular crystallization.

The different facets of the utilization of oil demonstrate that an individual oil and/or combinations of different oils can influence the outcome of crystallization experiments. The oil can play a part in the control of nucleation, affect the rate of equilibration and consequently determine the size of the forming crystals. Whether used for microbatch, vapour diffusion or for control of nucleation, the presence of oil is a parameter that can contribute to the accuracy, cleanliness and to the increase in the reproducibility of the experiments. Furthermore, the oil has a role in the protection of the trials during the course of their duration and in maintaining the stability of the resulting crystals.

Characterization of three distinct extradiol dioxygenases involved in mineralization of dibenzofuran by Terrabacter sp. strain DPO360.

The dibenzofuran-degrading bacterial strain DPO360 represents a new species of the genus Terrabacter together with the previously described dibenzofuran-mineralizing bacterial strain DPO1361 (K.-H. Engesser, V. Strubel, K. Christoglou, P. Fischer, and H. G. Rast, FEMS Microbiol. Lett. 65:205-210, 1989; V. Strubel, Ph.D. thesis, University of Stuttgart, Stuttgart, Germany, 1991; V. Strubel, H. G. Rast, W. Fietz, H.-J. Knackmuss, and K.-H. Engesser, FEMS Microbiol. Lett. 58:233-238, 1989). Two 2,3-dihydroxybiphenyl-1,2-dioxygenases (BphC1 and BphC2) and one catechol-2,3-dioxygenase (C23O) were shown to be expressed in Terrabacter sp. strain DPO360 growing with dibenzofuran as a sole source of carbon and energy. These enzymes exhibited strong sensitivity to oxygen. They were purified to apparent homogeneity as homodimers (BphC and BphC2) and as a homotetrameric catechol-2,3-dioxygenase (C23O). According to their specificity constants kcat/Km, both BphC1 and BphC2 were shown to be responsible for the cleavage of 2,2',3-trihydroxybiphenyl, the first metabolite in dibenzofuran mineralization along the angular dioxygenation pathway. With this substrate, BphC2 exhibited a considerably higher kcat/Km, value (183 microM/min) than BphC1 (29 microM/min). Catechol-2,3-dioxygenase was recognized to be not involved in the ring cleavage of 2,2',3-trihydroxybiphenyl (kcat/Km, 1 microM/min). Analysis of deduced amino acid sequence data of bphC1 revealed 36% sequence identity to nahC from Pseudomonas putida PpG7 (S. Harayama and M. Rekik, J. Biol. Chem. 264:15328-15333, 1989) and about 40% sequence identity to various bphC genes from different Pseudomonas and Rhodococcus strains. In addition, another 2,3-dihydroxybiphenyl-1,2-dioxygenase gene (bphC3) was cloned from the genome of Terrabacter sp. strain DPO360. Expression of this gene, however, could not be detected in Terrabacter sp. strain DPO360 after growth with dibenzofuran.

Mechanistic consequences of mutation of active site carboxylates in a retaining beta-1,4-glycanase from Cellulomonas fimi.

The exoglucanase/xylanase Cex from Cellulomonas fimi is a retaining glycosidase which functions via a two-step mechanism involving the formation and hydrolysis of a covalent glycosyl-enzyme intermediate. The roles of three conserved active site carboxylic acids in this enzyme have been probed by detailed kinetic analysis of mutants modified at these three positions. Elimination of the catalytic nucleophile (E233A) results in an essentially inactive enzyme, consistent with the important role of this residue. However addition of small anions such as azide or formate restores activity, but as an inverting enzyme since the product formed under these conditions is the alpha-glycosyl azide. Shortening of the catalytic nucleophile (E233D) reduces the rates of both formation and hydrolysis of the glycosyl-enzyme intermediate some 3000-4000-fold. Elimination of the acid/base catalyst (E127A) yields a mutant for which the deglycosylation step is slowed some 200-300-fold as a consequence of removal of general base catalysis, but with little effect on the transition state structure at the anomeric center. Effects on the glycosylation step due to removal of the acid catalyst depend on the aglycon leaving group ability, with minimal effects on substrates requiring no general acid catalysis but large (> 10(5)-fold) effects on substrates with poor leaving groups. The Brønsted beta 1g value for hydrolysis of aryl cellobiosides was much larger (beta 1g approximately -1) for the mutant than for the wild-type enzyme (beta 1g = -0.3), consistent with removal of protonic assistance. The pH-dependence was also significantly perturbed. Mutation of a third conserved active site carboxylic acid (E123A) resulted in rate reductions of up to 1500-fold on poorer substrates, which could be largely restored by addition of azide, but without the formation of glycosyl azide products. These results suggest a simple strategy for the identification of the key active site nucleophile and acid/base catalyst residues in glycosidases without resort to active site labeling.