Repetitive architecture of the Haemophilus influenzae Hia trimeric autotransporter

The Hia autotransporter of Haemophilus influenzae belongs to the trimeric autotransporter subfamily and mediates bacterial adherence to the respiratory epithelium. In this report, we show that the structure of Hia is characterized by a modular architecture containing repeats of structurally distinct domains. Comparison of the structures of HiaBD1 and HiaBD2 adhesive repeats and a nonadhesive repeat (a novel fold) shed light on the structural determinants of Hia adhesive function. Examination of the structure of an extended version of the Hia translocator domain revealed the structural transition between the C-terminal translocator domain and the N-terminal passenger domain, highlighting a highly intertwined domain that is ubiquitous among trimeric autotransporters. Overall, this study provides important insights into the mechanism of Hia adhesive activity and the overall structure of trimeric autotransporters.     

Structure of Bacteriophage T4 Endonuclease II Mutant E118A, a Tetrameric GIY-YIG Enzyme

Coliphage T4 endonuclease II (EndoII), encoded by gene denA, is a small (16 kDa, 136 aa) enzyme belonging to the GIY-YIG family of endonucleases, which lacks a C-terminal domain corresponding to that providing most of the binding energy in the structurally characterized GIY-YIG endonucleases, I-TevI and UvrC. In vivo, it is involved in degradation of host DNA, permitting scavenging of host-derived nucleotides for phage DNA synthesis. EndoII primarily catalyzes single-stranded nicking of DNA; 5- to 10-fold less frequently double-stranded breaks are produced. The Glu118Ala mutant of EndoII was crystallized in space group P2₁ with four monomers in the asymmetric unit. The fold of the EndoII monomer is similar to that of the catalytic domains of UvrC and I-TevI. In contrast to these enzymes, EndoII forms a striking X-shaped tetrameric structure composed as a dimer of dimers, with a protruding hairpin domain not present in UvrC or I-TevI providing most of the dimerization and tetramerization interfaces. A bound phosphate ion in one of the four active sites of EndoII likely mimics the scissile phosphate in a true substrate complex. In silico docking experiments showed that a protruding loop containing a nuclease-associated modular domain 3 element is likely to be involved in substrate binding, as well as residues forming a separate nucleic acid binding surface adjacent to the active site. The positioning of these sites within the EndoII primary dimer suggests that the substrate would bind to a primary EndoII dimer diagonally over the active sites, requiring significant distortion of the enzyme or the substrate DNA, or both, for simultaneous nicking of both DNA strands. The scarcity of potential nucleic acid binding residues between the active sites indicates that EndoII may bind its substrate inefficiently across the two sites in the dimer, offering a plausible explanation for the catalytic preponderance of single-strand nicks. Mutations analyzed in earlier functional studies are discussed in their structural context.     

Substitution of Glu122 by Glutamine Revealed the Function of the Second Water Molecule as a Proton Donor in the Binuclear Metal Enzyme Creatininase

Creatininase is a binuclear zinc enzyme and catalyzes the reversible conversion of creatinine to creatine. It exhibits an open-closed conformational change upon substrate binding, and the differences in the conformations of Tyr121, Trp154, and the loop region containing Trp174 were evident in the enzyme-creatine complex when compared to those in the ligand-free enzyme. We have determined the crystal structure of the enzyme complexed with a 1-methylguanidine. All subunits in the complex existed as the closed form, and the binding mode of creatinine was estimated. Site-directed mutagenesis revealed that the hydrophobic residues that show conformational change upon substrate binding are important for the enzyme activity.We propose a catalytic mechanism of creatininase in which two water molecules have significant roles. The first molecule is a hydroxide ion (Wat1) that is bound as a bridge between the two metal ions and attacks the carbonyl carbon of the substrate. The second molecule is a water molecule (Wat2) that is bound to the carboxyl group of Glu122 and functions as a proton donor in catalysis. The activity of the E122Q mutant was very low and it was only partially restored by the addition of ZnCl₂ or MnCl₂. In the E122Q mutant, k_cat is drastically decreased, indicating that Glu122 is important for catalysis. X-ray crystallographic study and the atomic absorption spectrometry analysis of the E122Q mutant-substrate complex revealed that the drastic decrease of the activity of the E122Q was caused by not only the loss of one Zn ion at the Metal1 site but also a critical function of Glu122, which most likely exists for a proton transfer step through Wat2.     

The capsid of the small RNA phage PRR1 is stabilized by metal ions

Many nonenveloped virus particles are stabilized by calcium ions bound in the interfaces between the protein subunits. These ions may have a role in the disassembly process. The small RNA phages of the Leviviridae family have T = 3 quasi-symmetry and are unique among simple viruses in that they have a coat protein with a translational repressor activity and a fold that has not been observed in other viruses. The crystal structure of phage PRR1 has been determined to 3.5 Å resolution. The structure shows a tentative binding site for a calcium ion close to the quasi-3-fold axis. The RNA-binding surface used for repressor activity is mostly conserved. The structure does not show any significant differences between quasi-equivalent subunits, which suggests that the assembly is not controlled by conformational switches as in many other simple viruses.     

Structure of the food-poisoning Clostridium perfringens enterotoxin reveals similarity to the aerolysin-like pore-forming toxins

Clostridium perfringens enterotoxin (CPE) is a major cause of food poisoning and antibiotic-associated diarrhea. Upon its release from C. perfringens spores, CPE binds to its receptor, claudin, at the tight junctions between the epithelial cells of the gut wall and subsequently forms pores in the cell membranes. A number of different complexes between CPE and claudin have been observed, and the process of pore formation has not been fully elucidated. We have determined the three-dimensional structure of the soluble form of CPE in two crystal forms by X-ray crystallography, to a resolution of 2.7 and 4.0 Å, respectively, and found that the N-terminal domain shows structural homology with the aerolysin-like β-pore-forming family of proteins. We show that CPE forms a trimer in both crystal forms and that this trimer is likely to be biologically relevant but is not the active pore form. We use these data to discuss models of pore formation.     

The Structure of Xis Reveals the Basis for Filament Formation and Insight into DNA Bending within a Mycobacteriophage Intasome

The recombination directionality factor, Xis, is a DNA bending protein that determines the outcome of integrase-mediated site-specific recombination by redesign of higher-order protein–DNA architectures. Although the attachment site DNA of mycobacteriophage Pukovnik is likely to contain four sites for Xis binding, Xis crystals contain five subunits in the asymmetric unit, four of which align into a Xis filament and a fifth that is generated by an unusual domain swap. Extensive intersubunit contacts stabilize a bent filament-like arrangement with Xis monomers aligned head to tail. The structure implies a DNA bend of ~ 120°, which is in agreement with DNA bending measured in vitro. Formation of attR-containing intasomes requires only Int and Xis, distinguishing Pukovnik from lambda. Therefore, we conclude that, in Pukovnik, Xis-induced DNA bending is sufficient to promote intramolecular Int-mediated bridges during intasome formation.     

The crystal structure of Cupriavidus necator nitrate reductase in oxidized and partially reduced states

The periplasmic nitrate reductase (NapAB) from Cupriavidus necator is a heterodimeric protein that belongs to the dimethyl sulfoxide reductase family of mononuclear Mo-containing enzymes and catalyzes the reduction of nitrate to nitrite. The protein comprises a large catalytic subunit (NapA, 91 kDa) containing the molybdenum active site plus one [4Fe-4S] cluster, as well as a small subunit (NapB, 17 kDa), which is a diheme c-type cytochrome involved in electron transfer. Crystals of the oxidized form of the enzyme diffracted beyond 1.5 Å at the European Synchrotron Radiation Facility. This is the highest resolution reported to date for a nitrate reductase, providing true atomic details of the protein active center, and this showed further evidence on the molybdenum coordination sphere, corroborating previous data on the related Desulfovibrio desulfuricans NapA. The molybdenum atom is bound to a total of six sulfur atoms, with no oxygen ligands or water molecules in the vicinity. In the present work, we were also able to prepare partially reduced crystals that revealed two alternate conformations of the Mo-coordinating cysteine. This crystal form was obtained by soaking dithionite into crystals grown in the presence of the ionic liquid [C₄mim]Cl⁻. In addition, UV-Vis and EPR spectroscopy studies showed that the periplasmic nitrate reductase from C. necator might work at unexpectedly high redox potentials when compared to all periplasmic nitrate reductases studied to date.     

Characterization of Two Novel Aldo-Keto Reductases from Arabidopsis: Expression Patterns, Broad Substrate Specificity, and an Open Active-Site Structure Suggest a Role in Toxicant Metabolism Following Stress

Aldo-keto reductases (AKRs) are widely distributed in nature and play numerous roles in the metabolism of steroids, sugars, and other carbonyls. They have also frequently been implicated in the metabolism of exogenous and endogenous toxicants, including those stimulated by stress. Although the Arabidopsis genome includes at least 21 genes with the AKR signature, very little is known of their functions. In this study, we have screened the Arabidopsis thaliana genomic sequence for genes with significant homology to members of the mammalian AKR1 family and identified four homologues for further study. Following alignment of the predicted protein sequences with representatives from the AKR superfamily, the proteins were ascribed not to the AKR1 family but to the AKR4C subfamily, with the individual designations of AKR4C8, AKR4C9, AKR4C10, and AKR4C11. Expression of two of the genes, AKR4C8 and AKR4C9, has been shown to be coordinately regulated and markedly induced by various forms of stress. The genes have been overexpressed in bacteria, and recombinant proteins have been purified and crystallized. Both enzymes display NADPH-dependent reduction of carbonyl compounds, typical of the superfamily, but will accept a very wide range of substrates, reducing a range of steroids, sugars, and aliphatic and aromatic aldehydes/ketones, although there are distinct differences between the two enzymes. We have obtained high-resolution crystal structures of AKR4C8 (1.4 Å) and AKR4C9 (1.25 Å) in ternary complexes with NADP⁺ and acetate. Three extended loops, present in all AKRs and responsible for defining the cofactor- and substrate-binding sites, are shorter in the 4C subfamily compared to other AKRs. Consequently, the crystal structures reveal open and accommodative substrate-binding sites, which correlates with their broad substrate specificity. It is suggested that the primary role of these enzymes may be to detoxify a range of toxic aldehydes and ketones produced during stress, although the precise nature of the principal natural substrates remains to be determined.     

Structural and Functional Analysis of SoPIP2;1 Mutants Adds Insight into Plant Aquaporin Gating

Plant plasma membrane aquaporins facilitate water flux into and out of plant cells, thus coupling their cellular function to basic aspects of plant physiology. Posttranslational modifications of conserved phosphorylation sites, changes in cytoplasmic pH and the binding of Ca²⁺ can regulate water transport activity by gating the plasma membrane aquaporins. A structural mechanism unifying these diverse biochemical signals has emerged for the spinach aquaporin SoPIP2;1, although several questions concerning the opening mechanism remain. Here, we describe the X-ray structures of the S115E and S274E single SoPIP2;1 mutants and the corresponding double mutant. Phosphorylation of these serines is believed to increase water transport activity of SoPIP2;1 by opening the channel. However, all mutants crystallised in a closed conformation, as confirmed by water transport assays, implying that neither substitution fully mimics the phosphorylated state. Nevertheless, a half-turn extension of transmembrane helix 1 occurs upon the substitution of Ser115, which draws the C^α atom of Glu31 10 Å away from its wild-type conformation, thereby disrupting the divalent cation binding site involved in the gating mechanism. Mutation of Ser274 disorders the C-terminus but no other significant conformational changes are observed. Inspection of the hydrogen-bond interactions within loop D suggested that the phosphorylation of Ser188 may also produce an open channel, and this was supported by an increased water transport activity for the S188E mutant and molecular dynamics simulations. These findings add additional insight into the general mechanism of plant aquaporin gating.     

Crystal structures of penicillin-binding protein 3 from Pseudomonas aeruginosa: comparison of native and antibiotic-bound forms

We report the first crystal structures of a penicillin-binding protein (PBP), PBP3, from Pseudomonas aeruginosa in native form and covalently linked to two important β-lactam antibiotics, carbenicillin and ceftazidime. Overall, the structures of apo and acyl complexes are very similar; however, variations in the orientation of the amino-terminal membrane-proximal domain relative to that of the carboxy-terminal transpeptidase domain indicate interdomain flexibility. Binding of either carbenicillin or ceftazidime to purified PBP3 increases the thermostability of the enzyme significantly and is associated with local conformational changes, which lead to a narrowing of the substrate-binding cleft. The orientations of the two β-lactams in the active site and the key interactions formed between the ligands and PBP3 are similar despite differences in the two drugs, indicating a degree of flexibility in the binding site. The conserved binding mode of β-lactam-based inhibitors appears to extend to other PBPs, as suggested by a comparison of the PBP3/ceftazidime complex and the Escherichia coli PBP1b/ceftoxamine complex. Since P. aeruginosa is an important human pathogen, the structural data reveal the mode of action of the frontline antibiotic ceftazidime at the molecular level. Improved drugs to combat infections by P. aeruginosa and related Gram-negative bacteria are sought and our study provides templates to assist that process and allows us to discuss new ways of inhibiting PBPs.     

The extra-membranous domains of the competence protein HofQ show DNA binding, flexibility and a shared fold with type I KH domains

Secretins form large oligomeric assemblies in the membrane that control both macromolecular secretion and uptake. Several Pasteurellaceae are naturally competent for transformation, but the mechanism for DNA assimilation is largely unknown. In Haemophilus influenzae, the secretin ComE has been demonstrated to be essential for DNA uptake. In closely related Aggregatibacter actinomycetemcomitans, an opportunistic pathogen in periodontitis, the ComE homolog HofQ is believed to be the outer membrane DNA translocase. Here, we report the structure of the extra-membranous domains of HofQ at 2.3 Å resolution by X-ray crystallography. We also show that the extra-membranous domains of HofQ are capable of DNA binding. The structure reveals two secretin-like folds, the first of which is formed by means of a domain swap. The second domain displays extensive structural similarity to K homology (KH) domains, including the presence of a GxxG motif, which is essential for the nucleotide-binding function of KH domains, suggesting a possible mechanism for DNA binding by HofQ. The data indicate a direct involvement in DNA acquisition and provide insight into the molecular basis for natural competence.     

Active-site mobility revealed by the crystal structure of arylmalonate decarboxylase from Bordetella bronchiseptica

Arylmalonate decarboxylase (AMDase) from Bordetella bronchiseptica catalyzes the enantioselective decarboxylation of arylmethylmalonates without the need for an organic cofactor or metal ion. The decarboxylation reaction is of interest for the synthesis of fine chemicals. As basis for an analysis of the catalytic mechanism of AMDase and for a rational enzyme design, we determined the X-ray structure of the enzyme up to 1.9 Å resolution. Like the distantly related aspartate or glutamate racemases, AMDase has an aspartate transcarbamoylase fold consisting of two α/β domains related by a pseudo dyad. However, the domain orientation of AMDase differs by about 30° from that of the glutamate racemases, and also significant differences in active-site structures are observed. In the crystals, four independent subunits showing different conformations of active-site loops are present. This finding is likely to reflect the active-site mobility necessary for catalytic activity.     

Structure and Function of P19, a High-Affinity Iron Transporter of the Human Pathogen Campylobacter jejuni

Campylobacter jejuni, a major cause of acute bacterial diarrhea in humans, expresses numerous proteins to import diverse forms of essential iron. The expression of p19 and an adjacent iron transporter homologue (ftr1) is strongly induced upon iron limitation, suggesting a function in iron acquisition. Here, we show that the loss of P19 alone is detrimental to growth on iron-restricted media. Furthermore, metal binding analysis demonstrates that recombinant P19 has distinct copper and iron binding sites. Crystal structures of P19 have been solved to 1.41 Å resolution, revealing an immunoglobulin-like fold. A P19 homodimer in which both monomers contribute ligands to two equivalent copper sites located adjacent to methionine-rich patches is observed. Copper coordination occurs via three histidine residues (His42, His95, and His132) and Met88. A solvent channel lined with conserved acidic residues leads to the copper site. Soaking crystals with a solution of manganese as iron analog reveals a second metal binding site in this solvent channel (metal-metal distance, 7.7 Å). Glu44 lies between the metal sites and displays multiple conformations in the crystal structures, suggesting a role in regulating metal-metal interaction. Dimerization is shown to be metal dependent in vitro and is detected in vivo by cross-linking.     

Crystal Structure of the ATP-Dependent Maturation Factor of Ni,Fe-Containing Carbon Monoxide Dehydrogenases

CooC proteins are ATPases involved in the incorporation of nickel into the complex active site ([Ni-4Fe-4S]) cluster of Ni,Fe-dependent carbon monoxide dehydrogenases. The genome of the carboxydotrophic bacterium Carboxydothermus hydrogenoformans encodes five carbon monoxide dehydrogenases and three CooC-type proteins, of which CooC1 was shown to be a nickel-binding ATPase. We determined the crystal structure of CooC1 in four different states: empty, ADP-bound, Zn²⁺/ADP-bound, and Zn²⁺-bound. The structure of CooC1 consists of two spatially separated functional modules: an ATPase module containing the deviant Walker A motif and a metal-binding module that confers the specific function of CooC1. The ATPase module is homologous to other members of the MinD family and, in analogy to the dimeric structure of ATP-bound Soj, is likely responsible for the ATP-dependent dimerization of CooC1. Its core topology classifies CooC1 as a member of the MinD family of SIMIBI (signal recognition particle, MinD and BioD)-class NTPases. The crystal structure of Zn²⁺-bound CooC1 reveals a conserved C-X-C motif as the metal-binding site responsible for metal-induced dimerization. The competitive binding of Ni²⁺ and Zn²⁺ to CooC1 in solution confirms that the conserved C-X-C motif is also responsible for the interaction with Ni²⁺. A comparison of the different CooC1 structures determined suggests a mutual dependence of metal-binding site and nucleotide-binding site.     

Structure of the Cytosolic Portion of the Motor Protein Prestin and Functional Role of the STAS Domain in SLC26/SulP Anion Transporters

Prestin is the motor protein responsible for the somatic electromotility of cochlear outer hair cells and is essential for normal hearing sensitivity and frequency selectivity of mammals. Prestin is a member of mammalian solute-linked carrier 26 (SLC26) anion exchangers, a family of membrane proteins capable of transporting a wide variety of monovalent and divalent anions. SLC26 transporters play important roles in normal human physiology in different tissues, and many of them are involved in genetic diseases. SLC26 and related SulP transporters carry a hydrophobic membrane core and a C-terminal cytosolic portion that is essential in plasma membrane targeting and protein function. This C-terminal portion is mainly composed of a STAS (sulfate transporters and anti-sigma factor antagonist) domain, whose name is due to a remote but significant sequence similarity with bacterial ASA (anti-sigma factor antagonist) proteins. Here we present the crystal structure at 1.57 Å resolution of the cytosolic portion of prestin, the first structure of a SulP transporter STAS domain, and its characterization in solution by heteronuclear multidimensional NMR spectroscopy. Prestin STAS significantly deviates from the related bacterial ASA proteins, especially in the N-terminal region, which—although previously considered merely as a generic linker between the domain and the last transmembrane helix—is indeed fully part of the domain. Hence, unexpectedly, our data reveal that the STAS domain starts immediately after the last transmembrane segment and lies beneath the lipid bilayer. A structure-function analysis suggests that this model can be a general template for most SLC26 and SulP anion transporters and supports the notion that STAS domains are involved in functionally important intramolecular and intermolecular interactions. Mapping of disease-associated or functionally harmful mutations on STAS structure indicates that they can be divided into two categories: those causing significant misfolding of the domain and those altering its interaction properties.     

Crystal structure of inulosucrase from Lactobacillus: insights into the substrate specificity and product specificity of GH68 fructansucrases

Fructansucrases (FSs) catalyze a transfructosylation reaction with sucrose as substrate to produce fructo-oligosaccharides and fructan polymers that contain either β-2,1 glycosidic linkages (inulin) or β-2,6 linkages (levan). Levan-synthesizing FSs (levansucrases) have been most extensively investigated, while detailed information on inulosucrases is limited. Importantly, the molecular basis of the different product specificities of levansucrases and inulosucrases is poorly understood.We have elucidated the three-dimensional structure of a truncated active bacterial GH68 inulosucrase, InuJ of Lactobacillus johnsonii NCC533 (residues 145-708), in its apo form, with a bound substrate (sucrose), and with a transfructosylation product. The sucrose binding pocket and the sucrose binding mode are virtually identical with those of GH68 levansucrases, confirming that both enzyme types use the same fully conserved structural framework for the binding and cleavage of the donor substrate sucrose in the active site. The binding mode of the first transfructosylation product 1-kestose (Fru-β(2-1)-Fru-α(2-1)-Glc, where Fru = fructose and Glc = glucose) in subsites − 1 to + 2 shows for the first time how inulin-type fructo-oligosaccharide bind in GH68 FS and how an inulin-type linkage can be formed. Surprisingly, observed interactions with the sugar in subsites + 1 and + 2 are provided by residues that are also present in levansucrases. The binding mode of 1-kestose and the presence of a more distant sucrose binding site suggest that residues beyond the + 2 subsite, in particular residues from the nonconserved 1B-1C loop, determine product linkage type specificity in GH68 FSs.