The structural basis for the exo-mode of action in GH74 oligoxyloglucan reducing end-specific cellobiohydrolase

Oligoxyloglucan reducing end-specific cellobiohydrolase (OXG-RCBH) is a unique exo-beta-1,4-glucanase that belongs to glycoside hydrolase family 74. The enzyme recognizes the reducing end of xyloglucan oligosaccharides and releases two glucosyl residue segments from the reducing end of the main chain. Previously, we reported that OXG-RCBH consists of two seven-bladed beta-propeller domains. There is a large cleft between the two domains, and a unique loop encloses one side of the active site cleft. Here, we report the X-ray crystal structure of the OXG-RCBH-substrate complex determined to a resolution of 2.4 A. The substrate bound to the cleft, and its reducing end was arranged near the loop region that is believed to impart OXG-RCBH with its activity. We constructed a deletion mutant of the loop region and conducted a detailed analysis. A deletion mutant of the loop region showed endo-activity with altered substrate recognition. More specifically, cleavage occurred randomly instead of at specific sites, most likely due to the misalignment of the substrate within the subsite. We believe that the loop imparts unique substrate specificity with exo-mode hydrolysis in OXG-RCBH.     

Purification,characterization, cloning,and expression of a novel xyloglucan-specific glycosidase,oligoxyloglucan reducing end-specific cellobiohydrolase

A novel oligoxyloglucan-specific glycosidase, oligoxyloglucan reducing end-specific cellobiohydrolase (OXG-RCBH), with a molecular mass of 97 kDa and a pI of 6.1, was isolated from the fungus Geotrichum sp. M128. Analysis of substrate specificity using various xyloglucan oligosaccharide structures revealed that OXG-RCBH had exoglucanase activity. It recognized the reducing end of oligoxyloglucan and released two glucosyl residue segments from the main chain. The full-length cDNA encoding OXG-RCBH was cloned and sequenced, and it had a 2436-bp open reading frame encoding an 812amino acid protein. The deduced protein showed approximately 35% identity to members of glycoside hydrolase family 74. The cDNA encoding OXG-RCBH was then expressed in Escherichia coli. Although the recombinant protein was expressed as an inclusion body, renaturation was successful, and enzymatically active recombinant OXG-RCBH was obtained.     

Functional and topological analysis of the Burkholderia cenocepacia priming glucosyltransferase BceB, involved in the biosynthesis of the cepacian exopolysaccharide

The BceB protein of the cystic fibrosis mucoid isolate Burkholderia cenocepacia IST432 is proposed to catalyze the first step of the exopolysaccharide repeat unit assembly. Extracts of Escherichia coli cells overexpressing BceB were shown to contain glycosyltransferase activity and mediate incorporation of glucose-1-phosphate into membrane lipids. The amino acid sequence of BceB exhibits two conserved regions, one comprising two invariant aspartic acid residues (Asp339 and Asp355) that are essential for catalysis, as substantiated by site-directed mutagenesis, and the other comprising a putative Rossmann fold motif. The results of protein topology analysis using PhoA and LacZ fusions supported in silico predictions that BceB has at least six transmembrane segments and two major cytoplasmic loops comprising the conserved regions described above.     

Drosophila substrate adhesion molecule: sequence of laminin B1 chain reveals domains of homology with mouse

Laminin, a substrate adhesion molecule in vertebrates, is a large glycoprotein complex in basement membranes that promotes cell adhesion, cell migration, and neurite outgrowth. Here we report on the cloning of the genes encoding the three subunits of Drosophila laminin. Sequence analysis of cDNA clones encoding the Drosophila B1 chain reveals a multidomain structure similar to that of its mouse homolog. The Drosophila sequence has only 25% amino acid identity with the mouse sequence in domains I, II, and IV. However, in one of the putative collagen-binding regions (domain VI) and the two cysteine-rich domains of EGF-like repeats (domains III and V), the amino acid identity between these two evolutionarily distant species jumps to 55%. Moreover, the number, length, and unique amino acid sequences of each of the 13 EGF-like repeats are highly conserved between Drosophila and mouse, suggesting that each may serve a unique function in protein-protein interactions.     

Crystal structure of haloalkane dehalogenase: an enzyme to detoxify halogenated alkanes.

Haloalkane dehalogenase from Xanthobacter autotrophicus GJ10 converts 1-haloalkanes to the corresponding alcohols and halide ions with water as the sole cosubstrate and without any need for oxygen or cofactors. The three-dimensional structure has been determined by multiple isomorphous replacement techniques using three heavy atom derivatives. The structure has been refined at 2.4 A resolution to an R-factor of 17.9%. The monomeric enzyme is a spherical molecule and is composed to two domains: domain I has an alpha/beta type structure with a central eight-stranded mainly parallel beta-sheet. Domain II lies like a cap on top of domain I and consists of alpha-helices connected by loops. Except for the cap domain the structure resembles that of the dienelactone hydrolase in spite of any significant sequence homology. The putative active site is completely buried in an internal hydrophobic cavity which is located between the two domains. From the analysis of the structure it is suggested that Asp124 is the nucleophilic residue essential for the catalysis. It interacts with His289 which is hydrogen-bonded to Asp260.     

Crystal structures of the lytic transglycosylase MltA from N.gonorrhoeae and E.coli: insights into interdomain movements and substrate binding

MltA is a lytic transglycosylase of Gram-negative bacteria that cleaves the beta-1,4 glycosidic linkages between N-acetylmuramic acid (MurNAc) and N-acetylglucosamine (GlcNAc) in peptidoglycan. We have determined the crystal structures of MltA from Neisseria gonorrhoeae and Escherichia coli (NgMltA and EcMltA), which have only 21.5% sequence identity. Both proteins have two main domains separated by a deep groove. Domain 1 shows structural similarity with the so-called double-psi barrel family of proteins. Comparison of the two structures reveals substantial differences in the relative positions of domains 1 and 2 such that the active site groove in NgMltA is much wider and appears more able to accommodate peptidoglycan substrate than EcMltA, suggesting that domain closure occurs after substrate binding. Docking of a peptidoglycan molecule into the structure of NgMltA reveals a number of conserved residues that are likely involved in substrate binding, including a potential binding pocket for the peptidyl moieties. This structure supports the assignment of Asp405 as the acid catalyst responsible for cleavage of the glycosidic bond. In EcMltA, the equivalent residue is Asp328, which has been identified previously. The structures also suggest a catalytic role for Asp393 (Asp317 in EcMltA) in activating the C6 hydroxyl group during formation of the 1,6-anhydro linkage. Finally, in comparison to EcMltA, NgMltA contains a unique third domain that is an insertion within domain 2. The domain is beta in structure and may mediate protein-protein interactions that are specific to peptidoglycan metabolism in N.gonorrhoeae.     

Mutagenesis of two acidic active site residues in human muscle creatine kinase: implications for the catalytic mechanism

Creatine kinase (CK) catalyzes the reversible phosphorylation of the guanidine substrate, creatine, by MgATP. Although several X-ray crystal structures of various isoforms of creatine kinase have been published, the detailed catalytic mechanism remains unresolved. A crystal structure of the CK homologue, arginine kinase (AK), complexed with the transition-state analogue (arginine-nitrate-ADP), has revealed two carboxylate amino acid residues (Glu225 and Glu314) within 2.8 A of the proposed transphosphorylation site. These two residues are the putative catalytic groups that may promote nucleophilic attack by the guanidine amino group on the gamma-phosphate of ATP. From primary sequence alignments of arginine kinases and creatine kinases, we have identified two homologous creatine kinase acidic amino acid residues (Glu232 and Asp326), and these were targeted for examination of their potential roles in the CK mechanism. Using site-directed mutagenesis, we have made several substitutions at these two positions. The results indicate that of these two residues the Glu232 is the likely catalytic residue while Asp326 likely performs a role in properly aligning substrates for catalysis.     

Structural basis for the activation of phenylalanine in the non-ribosomal biosynthesis of gramicidin S.

The non-ribosomal synthesis of the cyclic peptide antibiotic gramicidin S is accomplished by two large multifunctional enzymes, the peptide synthetases 1 and 2. The enzyme complex contains five conserved subunits of approximately 60 kDa which carry out ATP-dependent activation of specific amino acids and share extensive regions of sequence similarity with adenylating enzymes such as firefly luciferases and acyl-CoA ligases. We have determined the crystal structure of the N-terminal adenylation subunit in a complex with AMP and L-phenylalanine to 1.9 A resolution. The 556 amino acid residue fragment is folded into two domains with the active site situated at their interface. Each domain of the enzyme has a similar topology to the corresponding domain of unliganded firefly luciferase, but a remarkable relative domain rotation of 94 degrees occurs. This conformation places the absolutely conserved Lys517 in a position to form electrostatic interactions with both ligands. The AMP is bound with the phosphate moiety interacting with Lys517 and the hydroxyl groups of the ribose forming hydrogen bonds with Asp413. The phenylalanine substrate binds in a hydrophobic pocket with the carboxylate group interacting with Lys517 and the alpha-amino group with Asp235. The structure reveals the role of the invariant residues within the superfamily of adenylate-forming enzymes and indicates a conserved mechanism of nucleotide binding and substrate activation.     

The Complete Primary Structure of Molluscan Shell Protein 1 (MSP-1), an Acidic Glycoprotein in the Shell Matrix of the Scallop Patinopecten yessoensis

The complete primary structure of MSP-1, a major water-soluble glycoprotein in the foliated calcite shell layer of the scallop Patinopecten yessoensis, is reported. The full-length complementary DNA for MSP-1 isolated by polymerase chain reaction contained a sequence for a signal peptide of 20 amino acids followed by a polypeptide of 820 amino acids with calculated molecular mass of 74.5 kDa. The deduced amino acid sequence of MSP-1 includes a high proportion of Ser (32%), Gly (25%), and Asp (20%), and the predicted isoelectric point is 3.2; in these respects, MSP-1 is a typical acidic glycoprotein of mineralized tissues. A repeated modular structure characterizes MSP-1, with a sequence unit between 158 and 177 amino acids in length being repeated 4 times in tandem in the middle part of the protein. The repeated unit comprises 3 modules (SG, D, and K domains), each having a distinct amino acid composition and sequence. The SG domain is almost exclusively composed of Ser and Gly residues. The D domain is rich in Asp residues, potential N-glycosylation and phosphorylation sites. The K domain is rich in Gly residues and has a core of basic residues. The Asp residues are arranged more or less regularly in the D domains, exhibiting some repeated motifs such as Asp-Gly-Ser-Asp and Asp-Ser-Asp. Further, the 4 D domains indicate remarkable overall sequence similarities to each other. These observations suggest that the regular arrangements of COO⁻ groups in the D domain side chains may be important for specific control of crystal growth. Received September 19, 2000; accepted February 9, 2001     

L—乳酸脱氢酶基因克隆及功能分析

构建了一株产D,L-乳酸的乳杆菌(Lactobaeillus sp．)MD—1的基因库。利用乳酸脱氢酶和丙酮酸裂解酶缺陷的Escherichia coli FMJ144作为宿主,通过互补筛选分离克隆到乳酸脱氢酶基因(ldhL)。核酸序列分析表明,该基因以ATG为起始密码子编码316个氨基酸残基组成的蛋白质,预测的分子量为33．84kD;5′端存在典型的启动子结构,3′端的终止子是不依赖于ρ因子的转录终止子。ldhL编码的蛋白质有3个保守区域,其中Gly13～Asp50保守区域是NADH的结合位点,Asp73～Ile100和Asn123～Arg154保守区是酶的活性部位。该ldhL和其他乳杆菌的ldhL基因和编码的氨基酸序列相似性较低,核苷酸序列相似性最高仅为64．1％,氨基酸序列相似性最高仅为68．9％,是新的L—乳酸脱氢酶基因。     

The predicted structure of photopexin from Photorhabdus shows the first haemopexin-like motif in prokaryotes

The insect pathogenic bacterium Photorhabdus luminescens secretes several insecticidal high molecular mass 'toxin complexes'. Analysis of the putative pathogenicity island surrounding the toxin complex a (tca) locus revealed two open reading frames (ORFs) of unknown function. The predicted protein sequences of these ORFs show a repeated motif similar to those found in the vertebrate haem scavenging molecule haemopexin, limunectin (a phosphocholine binding protein from Limulus) and the C-terminal domains of matrix metalloproteinases (MMPs) (where they are thought to be important for cell attachment and adhesion). We have therefore named the operon photopexin AB and the putative encoded proteins 'photopexins' A and B (PpxA and PpxB). The predicted amino acid sequence of PpxA was modelled onto the crystal structure of a MMP. Our model predicts not only that PpxA and PpxB have beta-propeller domains but also that each haemopexin-like repeat corresponds to one blade of a propeller, suggesting the limunectin structure itself may also contain two or three such haemopexin-like propellers. The overall structure of PpxA has striking similarity to that of haemopexin suggesting that it may be used by the bacterium to scavenge iron containing compounds from insects. The implications for the potential role of Ppx proteins in pathogenicity are discussed. This is the first finding of a haemopexin-like repeat outside plants and animals.     

The structure of cell adhesion molecule uvomorulin. Insights into the molecular mechanism of Ca2+-dependent cell adhesion.

We have determined the amino acid sequence of the Ca2+-dependent cell adhesion molecule uvomorulin as it appears on the cell surface. The extracellular part of the molecule exhibits three internally repeated domains of 112 residues which are most likely generated by gene duplication. Each of the repeated domains contains two highly conserved units which could represent putative Ca2+-binding sites. Secondary structure predictions suggest that the putative Ca2+-binding units are located in external loops at the surface of the protein. The protein sequence exhibits a single membrane-spanning region and a cytoplasmic domain. Sequence comparison reveals extensive homology to the chicken L-CAM. Both uvomorulin and L-CAM are identical in 65% of their entire amino acid sequence suggesting a common origin for both CAMs.     

Crystal structure of archaeal RNase HII: a homologue of human major RNase H

BACKGROUND: RNases H are present in all organisms and cleave RNAs in RNA/DNA hybrids. There are two major types of RNases H that have little similarity in sequence, size and specificity. The structure of RNase HI, the smaller enzyme and most abundant in bacteria, has been extensively studied. However, no structural information is available for the larger RNase H, which is most abundant in eukaryotes and archaea. Mammalian RNase H participates in DNA replication, removal of the Okazaki fragments and possibly DNA repair. RESULTS: The crystal structure of RNase HII from the hypothermophile Methanococcus jannaschii, which is homologous to mammalian RNase H, was solved using a multiwavelength anomalous dispersion (MAD) phasing method at 2 A resolution. The structure contains two compact domains. Despite the absence of sequence similarity, the large N-terminal domain shares a similar fold with the RNase HI of bacteria. The active site of RNase HII contains three aspartates: Asp7, Asp112 and Asp149. The nucleotide-binding site is located in the cleft between the N-terminal and C-terminal domains. CONCLUSIONS: Despite a lack of any detectable similarity in primary structure, RNase HII shares a similar structural domain with RNase HI, suggesting that the two classes of RNases H have a common catalytic mechanism and possibly a common evolutionary origin. The involvement of the unique C-terminal domain in substrate recognition explains the different reaction specificity observed between the two classes of RNase H.     

Modelling,substrate docking and mutational analysis identify residues essential for function and specificity of the major fungal purine transporter AzgA

The AzgA purine/H⁺ symporter of Aspergillus nidulans is the founding member of a functionally and phylogenetically distinct transporter family present in fungi, bacteria and plants. Here a valid AzgA topological model is built based on the crystal structure of the Escherichia coli uracil transporter UraA, a member of the nucleobase‐ascorbate transporter (NAT/NCS2) family. The model consists of 14 transmembrane, mostly α‐helical, segments (TMSs) and cytoplasmic N‐ and C‐tails. A distinct compact core of 8 TMSs, made of two intertwined inverted repeats (TMSs 1–4 and 8–11), is topologically distinct from a flexible domain (TMSs 5–7 and 12–14). A putative substrate binding cavity is visible between the core and the gate domains. Substrate docking, molecular dynamics and mutational analysis identified several residues critical for purine binding and/or transport in TMS3, TMS8 and TMS10. Among these, Asn131 (TMS3), Asp339 (TMS8) and Glu394 (TMS10) are proposed to directly interact with substrates, while Asp342 (TMS8) might be involved in subsequent substrate translocation, through H⁺ binding and symport. Thus, AzgA and other NAT transporters use topologically similar TMSs and amino acid residues for substrate binding and transport, which in turn implies that AzgA‐like proteins constitute a distant subgroup of the ubiquitous NAT family.     

The complete primary structure for the alpha 1-chain of mouse collagen IV. Differential evolution of collagen IV domains

We report here the complete nucleotide and amino acid sequences for the alpha 1-chain of mouse collagen IV which is 1669 amino acids in length, including a putative 27-residue signal peptide. In comparison with the amino acid sequence for the alpha 2-chain (Saus, J., Quinones, S., MacKrell, A. J., Blumberg, B., Muthkumaran, G., Pihlajaniemi, J., and Kurkinen, M. (1989) J. Biol. Chem. 264, 6318-6324), the two chains of collagen IV are 43% identical. Most of the interruptions of the Gly-X-Y repeat are homologously placed but strikingly show no sequence similarity between the two chains. Availability of the amino acid sequences for human collagen IV allows a detailed comparison of the primary structure of collagen IV and reveals evolutionarily conserved domains of the protein. Between the two species, the alpha 1 (IV) chains are 90.6% and the alpha 2 (IV) chains are 83.5% identical in sequence. We discuss these data with respect to differential evolution between and within the collagen IV chain types.     

Identification and analysis of novel tandem repeats in the cell surface proteins of archaeal and bacterial genomes using computational tools

We have identified four novel repeats and two domains in cell surface proteins encoded by the Methanosarcina acetivorans genome and in some archaeal and bacterial genomes. The repeats correspond to a certain number of amino acid residues present in tandem in a protein sequence and each repeat is characterized by conserved sequence motifs. These correspond to: (a) a 42 amino acid (aa) residue RIVW repeat; (b) a 45 aa residue LGxL repeat; (c) a 42 aa residue LVIVD repeat; and (d) a 54 aa residue LGFP repeat. The domains correspond to a certain number of aa residues in a protein sequence that do not comprise internal repeats. These correspond to: (a) a 200 aa residue DNRLRE domain; and (b) a 70 aa residue PEGA domain. We discuss the occurrence of these repeats and domains in the different proteins and genomes analysed in this work.     

Crystal structure of Proteus vulgaris chondroitin sulfate ABC lyase I at 1.9A resolution

Chondroitin Sulfate ABC lyase I from Proteus vulgaris is an endolytic, broad-specificity glycosaminoglycan lyase, which degrades chondroitin, chondroitin-4-sulfate, dermatan sulfate, chondroitin-6-sulfate, and hyaluronan by beta-elimination of 1,4-hexosaminidic bond to unsaturated disaccharides and tetrasaccharides. Its structure revealed three domains. The N-terminal domain has a fold similar to that of carbohydrate-binding domains of xylanases and some lectins, the middle and C-terminal domains are similar to the structures of the two-domain chondroitin lyase AC and bacterial hyaluronidases. Although the middle domain shows a very low level of sequence identity with the catalytic domains of chondroitinase AC and hyaluronidase, the residues implicated in catalysis of the latter enzymes are present in chondroitinase ABC I. The substrate-binding site in chondroitinase ABC I is in a wide-open cleft, consistent with the endolytic action pattern of this enzyme. The tryptophan residues crucial for substrate binding in chondroitinase AC and hyaluronidases are lacking in chondroitinase ABC I. The structure of chondroitinase ABC I provides a framework for probing specific functions of active-site residues for understanding the remarkably broad specificity of this enzyme and perhaps engineering a desired specificity. The electron density map showed clearly that the deposited DNA sequence for residues 495-530 of chondroitin ABC lyase I, the segment containing two putative active-site residues, contains a frame-shift error resulting in an incorrectly translated amino acid sequence.     

Biochemical characterization and structural prediction of a novel cytosolic leucyl aminopeptidase of the M17 family from Schizosaccharomyces pombe

A new leucyl aminopeptidase activity has been identified in the fission yeast Schizosaccharomyces pombe. The enzyme, which has been purified and named leucyl aminopeptidase yspII (LAP yspII), had a molecular mass of 320 and 54 kDa by gel filtration and SDS/PAGE, respectively, suggesting a homohexameric structure. The enzyme cleaved synthetic aminoacyl-4-nitroanilides at an optimum of pH 8.5, and preferred leucine and methionine as N-terminal amino acids. A clear dependence on Mn2+ concentration for activity was found, and an apparent association constant of 0.33 mM was calculated for the metal ion. Bestatin behaved as a competitive inhibitor of LAP yspII (K(i) = 0.14 microM), while chelating agents such as chloroquine, EDTA and 1,10-phenanthroline also reduced enzyme activity. A MALDI-MS analysis, followed by sequencing of two of the resulting peptides, showed that LAP yspII undoubtedly corresponds to the putative aminopeptidase C13A11.05 identified in the S. pombe genome project. The protein exhibited nearly 40% sequence identity to fungal and mammalian aminopeptidases belonging to the M17 family of metallopeptidases. Catalytic residues (Lys292 and Arg366), as well as those involved in coordination with the cocatalytic metal ions (Lys280, Asp285, Asp303, Asp362 and Glu364) and those forming the hydrophobic pocket for substrate binding (Met300, Asn360, Ala363, Thr390, Leu391, Ala483 and Met486), were perfectly conserved among all known aminopeptidases. The S. pombe enzyme is predicted to be formed two clearly distinguished domains with a well conserved C-terminal catalytic domain showing a characteristic topology of eight beta-sheets surrounded by alpha-helical segments in the form of a saddle.