Crystal structure of a bicupin protein HutD involved in histidine utilization in Pseudomonas.

Cupins form one of the most functionally diverse superfamilies of proteins, with members performing a wide range of catalytic, non-catalytic, and regulatory functions. HutD is a predicted bicupin protein that is involved in histidine utilization (Hut) in Pseudomonas species. Previous genetic analyses have suggested that it limits the upper level of Hut pathway expression, but its mechanism of action is unknown. Here, we have determined the structure of PfluHutD at 1.74 Å resolution in several crystallization conditions, and identified N-formyl-l-glutamate (FG, a Hut pathway intermediate) as a potential ligand in vivo. Proteins 2017; 85:1580-1588. © 2017 Wiley Periodicals, Inc.

Total Chemical Synthesis of an Orf Virus Protein, ORFV002, an Inhibitor of the Master Gene Regulator NF-κB.

ORFV002 is a novel orf viral protein (117 Aa) that inhibits nuclear events through the regulation of the transcriptional activity of NF-κB, a master regulator of human gene expression (Diel et al., J Virol 2011, 85, 264-275). It is identified as the first nuclear inhibitor of NF-κB produced by orf virus (ORFV) and no homologues in other genera of the Chordopoxvirinae subfamily have been reported to date (Diel et al., J Virol 2011, 85, 264-275). Our molecular structure predictions suggest that ORFV002 may mimic part of IκB, an inhibitor and natural human partner of NF-κB. Recent advances in total chemical synthesis of proteins have provided solutions in overcoming challenges of current recombinant methods of protein isolation for structure elucidation. Aided by Boc solid phase peptide synthesis and native chemical ligation, ORFV002 was successfully synthesized in multimilligram amounts in good yield and high purity.

Pili in Gram-negative and Gram-positive bacteria - structure, assembly and their role in disease.

Many bacterial species possess long filamentous structures known as pili or fimbriae extending from their surfaces. Despite the diversity in pilus structure and biogenesis, pili in Gram-negative bacteria are typically formed by non-covalent homopolymerization of major pilus subunit proteins (pilins), which generates the pilus shaft. Additional pilins may be added to the fiber and often function as host cell adhesins. Some pili are also involved in biofilm formation, phage transduction, DNA uptake and a special form of bacterial cell movement, known as 'twitching motility'. In contrast, the more recently discovered pili in Gram-positive bacteria are formed by covalent polymerization of pilin subunits in a process that requires a dedicated sortase enzyme. Minor pilins are added to the fiber and play a major role in host cell colonization.This review gives an overview of the structure, assembly and function of the best-characterized pili of both Gram-negative and Gram-positive bacteria.

Protein adhesins as vaccine antigens for Group A Streptococcus.

Group A Streptococcus (GAS) is a globally important human pathogen that causes a broad spectrum of disease ranging from mild superficial infections to severe invasive diseases with high morbidity and mortality. Currently, there is no vaccine available for human use. GAS produces a vast array of virulence factors including multiple adhesin molecules. These mediate binding of the bacteria to host tissues and are essential in the initial phases of infection. Prophylactic vaccination with adhesins is a promising vaccine strategy and many GAS adhesins are currently in development as vaccine candidates. The most advanced candidates, having entered clinical trials, are based on the M protein, while components of the pilus and a number of fibronectin-binding proteins are in pre-clinical development. Adhesin-based vaccines aim to induce protective immunity via two main mechanisms: neutralisation where adhesin-specific antibodies block the ability of the adhesin to bind to host tissue and opsonisation in which adhesin-specific antibodies tag the GAS bacteria for phagocytosis. This review summarises our current knowledge of GAS adhesins and their structural features in the context of vaccine development.

Harnessing ester bond chemistry for protein ligation.

The hydrolysis potential of ester bonds in covalently cross-linking proteins is captured in our novel protein ligation technology. This new type of "molecular superglue" based on the spontaneously-formed Thr-Gln ester bonds found in cell-surface adhesins, affords a unique mechanism to both rationally assemble and disassemble complex protein nanomaterials.

Purification, crystallization and preliminary X-ray analysis of Mycobacterium tuberculosisfolylpolyglutamate synthase (MtbFPGS).

The gene encoding Mycobacterium tuberculosis FPGS (MtbFPGS; Rv2447c) has been cloned and the protein (51 kDa) expressed in Escherichia coli. The purified protein was crystallized either by the batch method in the presence of adenosine diphosphate (ADP) and CoCl2 or by vapour diffusion in the presence of ADP, dihydrofolate and CaCl2. X-ray diffraction data to approximately 2.0 and 2.6 A resolution were collected at the Stanford Synchrotron Radiation Laboratory (SSRL) for crystals grown under the respective conditions. Both crystals belong to the cubic space group P2(1)3, with a unit-cell parameter of 112.6 and 111.8 A, respectively. Structure determination is proceeding.

The structure of glucose-fructose oxidoreductase from Zymomonas mobilis: an osmoprotective periplasmic enzyme containing non-dissociable NADP.

BACKGROUND: The organism Zymomonas mobilis occurs naturally in sugar-rich environments. To protect the bacterium against osmotic shock, the periplasmic enzyme glucose-fructose oxidoreductase (GFOR) produces the compatible, solute sorbitol by reduction of fructose, coupled with the oxidation of glucose to gluconolactone. Hence, Z mobilis can tolerate high concentrations of sugars and this property may be useful in the development of an efficient microbial process for ethanol production. Each enzyme subunit contains tightly associated NADP which is not released during the catalytic cycle. RESULTS: The structure of GFOR was determined by X-ray crystallography at 2.7 A resolution. Each subunit of the tetrameric enzyme comprises two domains, a classical dinucleotide-binding domain, and a C-terminal domain based on a predominantly antiparallel nine-stranded beta sheet. In the tetramer, the subunits associate to form two extended 18-stranded beta sheets, which pack against each other in a face to face fashion, creating an extensive interface at the core of the tetramer. An N-terminal arm from each subunit wraps around the dinucleotide-binding domain of an adjacent subunit, covering the adenine ring of NADP. CONCLUSIONS: In GFOR, the NADP is found associated with a classical dinucleotide-binding domain in a conventional fashion. The NADP is effectively buried in the protein-subunit interior as a result of interactions with the N-terminal arm from an adjacent subunit in the tetramer, and with a short helix from the C-terminal domain of the protein. This accounts for NADP's inability to dissociate. The N-terminal arm may also contribute to stabilization of the tetramer. The enzyme has an unexpected structural similarity with the cytoplasmic enzyme glucose-6-phosphate dehydrogenase (G6PD). We hypothesize that both enzymes have diverged from a common ancestor. The mechanism of catalysis is still unclear, but we have identified a conserved structural motif (Glu-Lys-Pro) in the active site of GFOR and G6PD that may be important for catalysis.

The three-dimensional structure of PNGase F, a glycosylasparaginase from Flavobacterium meningosepticum.

BACKGROUND: Peptide:N-glycosidase F (PNGase F) is an enzyme that catalyzes the complete removal of N-linked oligosaccharide chains from glycoproteins. Often called an endoglycosidase, it is more correctly termed an amidase or glycosylasparaginase as cleavage is at the asparagine-sugar amide linkage. The enzyme is widely used in structure-function studies of glycoproteins. RESULTS: We have determined the crystal structure of PNGase F at 1.8 A resolution. The protein is folded into two domains, each with an eight-stranded antiparallel beta jelly roll configuration similar to many viral capsid proteins and also found, in expanded form, in lectins and several glucanases. Two potential active site regions have been identified, both in the interdomain region and shaped by prominent loops from one domain. Exposed aromatic residues are a feature of one site. CONCLUSIONS: The finding that PNGase F is based on two jelly roll domains suggests parallels with lectins and other carbohydrate-binding proteins. These proteins either bind sugars on the concave face of the beta-sandwich structure (aided by loops) or amongst the loops themselves. Further analysis of the function and identification of the catalytic site should lead to an understanding of both the specificity of PNGase F and possibly also the recognition processes that identify glycosylation sites on proteins.

1.8 A crystal structure of the C-terminal domain of rabbit serum haemopexin.

BACKGROUND: Haemopexin is a serum glycoprotein that binds haem reversibly and delivers it to the liver where it is taken up by receptor-mediated endocytosis. Haemopexin has two homologous domains, each having a characteristic fourfold internal sequence repeat. Haemopexin-type domains are also found in other proteins, including the serum adhesion protein vitronectin and various collagenases, in which they mediate protein-protein interactions. RESULTS: We have determined the crystal structure of the C-terminal domain of haemopexin at 1.8 A resolution. The domain is folded into four beta-leaflet modules, arranged in succession around a central pseudo-fourfold axis. A funnel-shaped tunnel through the centre of this disc-shaped domain serves as an ion-binding site. CONCLUSIONS: A model for haem binding by haemopexin is proposed, utilizing an anion-binding site at the wider end of the central tunnel, together with an associated cleft. This parallels the active-site location in other beta-propeller structures. The capacity to bind both cations and anions, together with the disc shape of the domain, suggests that such domains may be used widely for macromolecular recognition.

Crystal structure of methylmalonyl-coenzyme A epimerase from P. shermanii: a novel enzymatic function on an ancient metal binding scaffold.

BACKGROUND: Methylmalonyl-CoA epimerase (MMCE) is an essential enzyme in the breakdown of odd-numbered fatty acids and of the amino acids valine, isoleucine, and methionine. Present in many bacteria and in animals, it catalyzes the conversion of (2R)-methylmalonyl-CoA to (2S)-methylmalonyl-CoA, the substrate for the B12-dependent enzyme, methylmalonyl-CoA mutase. Defects in this pathway can result in severe acidosis and cause damage to the central nervous system in humans. RESULTS: The crystal structure of MMCE from Propionibacterium shermanii has been determined at 2.0 A resolution. The MMCE monomer is folded into two tandem betaalphabetabetabeta modules that pack edge-to-edge to generate an 8-stranded beta sheet. Two monomers then pack back-to-back to create a tightly associated dimer. In each monomer, the beta sheet curves around to create a deep cleft, in the floor of which His12, Gln65, His91, and Glu141 provide a binding site for a divalent metal ion, as shown by the binding of Co2+. Modeling 2-methylmalonate into the active site identifies two glutamate residues as the likely essential bases for the epimerization reaction. CONCLUSIONS: The betaalphabetabetabeta modules of MMCE correspond with those found in several other proteins, including bleomycin resistance protein, glyoxalase I, and a family of extradiol dioxygenases. Differences in connectivity are consistent with the evolution of these very different proteins from a common precursor by mechanisms of gene duplication and domain swapping. The metal binding residues also align precisely, and striking structural similarities between MMCE and glyoxalase I suggest common mechanisms in their respective epimerization and isomerization reactions.

Crystal structures of Bacillus caldovelox arginase in complex with substrate and inhibitors reveal new insights into activation, inhibition and catalysis in the arginase superfamily.

BACKGROUND: Arginase is a manganese-dependent enzyme that catalyzes the hydrolysis of L-arginine to L-ornithine and urea. In ureotelic animals arginase is the final enzyme of the urea cycle, but in many species it has a wider role controlling the use of arginine for other metabolic purposes, including the production of creatine, polyamines, proline and nitric oxide. Arginase activity is regulated by various small molecules, including the product L-ornithine. The aim of these structural studies was to test aspects of the catalytic mechanism and to investigate the structural basis of arginase inhibition. RESULTS: We report here the crystal structures of arginase from Bacillus caldovelox at pH 5.6 and pH 8.5, and of binary complexes of the enzyme with L-arginine, L-ornithine and L-lysine at pH 8.5. The arginase monomer comprises a single compact alpha/beta domain that further associates into a hexameric quaternary structure. The binary complexes reveal a common mode of ligand binding, which places the substrate adjacent to the dimanganese centre. We also observe a conformational change that impacts on the active site and is coupled with the occupancy of an external site by guanidine or arginine. CONCLUSIONS: The structures reported here clarify aspects of the active site and indicate key features of the catalytic mechanism, including substrate coordination to one of the manganese ions and an orientational role for a neighboring histidine residue. Stereospecificity for L-amino acids is found to depend on their precise recognition at the active-site rim. Identification of a second arginine-binding site, remote from the active site, and associated conformational changes lead us to propose a regulatory role for this site in substrate hydrolysis.

Sheep liver cytosolic aldehyde dehydrogenase: the structure reveals the basis for the retinal specificity of class 1 aldehyde dehydrogenases.

BACKGROUND: . Enzymes of the aldehyde dehydrogenase family are required for the clearance of potentially toxic aldehydes, and are essential for the production of key metabolic regulators. The cytosolic, or class 1, aldehyde dehydrogenase (ALDH1) of higher vertebrates has an enhanced specificity for all-trans retinal, oxidising it to the powerful differentiation factor all-trans retinoic acid. Thus, ALDH1 is very likely to have a key role in vertebrate development. RESULTS: . The three-dimensional structure of sheep ALDH1 has been determined by X-ray crystallography to 2.35 A resolution. The overall tertiary and quaternary structures are very similar to those of bovine mitochondrial ALDH (ALDH2), but there are important differences in the entrance tunnel for the substrate. In the ALDH1 structure, the sidechain of the general base Glu268 is disordered and the NAD+ cofactor binds in two distinct modes. CONCLUSIONS: . The submicromolar Km of ALDH1 for all-trans retinal, and its 600-fold enhanced affinity for retinal compared to acetaldehyde, are explained by the size and shape of the substrate entrance tunnel in ALDH1. All-trans retinal fits into the active-site pocket of ALDH1, but not into the pocket of ALDH2. Two helices and one surface loop that line the tunnel are likely to have a key role in defining substrate specificity in the wider ALDH family. The relative sizes of the tunnels also suggest why the bulky alcohol aversive drug disulfiram reacts more rapidly with ALDH1 than ALDH2. The disorder of Glu268 and the observation that NAD+ binds in two distinct modes indicate that flexibility is a key facet of the enzyme reaction mechanism.

The specificity of actinidin and its relationship to the structure of the enzyme.

The kinetic parameters kcat, Km and kcat/Km, have been determined for the actinidin-catalysed hydrolyses of N-substituted amino acid esters and amides and compared to the corresponding values for papain (EC 3.4.22.2). Substrates with aromatic N-substituents have lower kcat/Km values for actinidin (EC 3.4.22.14); the difference is much smaller for substrates with aliphatic substituents. The lower kcat/Km values for actinidin generally correspond to higher Km values suggesting that the strength of substrate binding differs between the two enzymes. This difference is explained in terms of the differences in the substrate binding sites found in X-ray crystallographic studies.

Structure of azurin from Alcaligenes denitrificans refinement at 1.8 A resolution and comparison of the two crystallographically independent molecules.

The structure of the blue copper protein azurin, from Alcaligenes denitrificans, has been refined crystallographically by restrained least-squares methods. The final crystallographic R value for 21,980 observed reflections to 1.8 A (1 A = 0.1 nm) resolution is 0.157. The asymmetric unit of the crystal contains two independent azurin molecules, the model for which comprises 1973 protein atoms, together with three SO2-4 ions, and 281 water molecules. Comparison of the two molecules shows very high correspondence. For 125 out of 129 residues (excluding only the chain termini, residues 1 to 2 and 128 to 129) the root-mean-square (r.m.s.) deviation in main-chain atom positions is 0.27 A. For other structural parameters r.m.s. deviations are also low; torsion angles 6.5 degrees, hydrogen bond lengths 0.12 A, bonds to copper 0.04 A and bond angles at the copper 3.9 degrees. The only significant differences are at the chain termini and in several loops. Some of these can be attributed to crystal packing effects, others to genuine structural microheterogeneity. Refinement has confirmed that the copper co-ordination is best described as distorted trigonal planar, with strong in-plane bonds to His46 N delta 1, His117 N delta 1 and Cys112 S gamma, and much weaker axial interactions with Met121 S delta and Gly45 C = O. Two N-H...S hydrogen bonds characterize Cys112 S gamma as a thiolate (S-) sulphur and may influence the visible absorption maximum. Atoms in and around the copper site have very low mobility, whereas the most mobile regions of the molecule are the chain termini and some of the connecting loops between secondary structure elements, especially those at the "southern" end, remote from the copper site. Main-chain to side-chain hydrogen bonds supply important stabilizing interactions at the "northern" end. Surface features include the hydrophobic patch around His117, probably important for electron transfer, the SO2-4 site at His83, and the general absence of ion pairs, despite the presence of many charged amino acid residues. The 281 water molecules include 182 that occur as approximately twofold-related pairs. There are no internal water molecules. The water sites common to both azurin molecules include those in surface pockets and some in intermolecular contact regions. They are characterized by relatively low thermal parameters and numerous protein contacts.

Thiol proteases. Comparative studies based on the high-resolution structures of papain and actinidin, and on amino acid sequence information for cathepsins B and H, and stem bromelain.

An accurate three-dimensional structure is known for papain (1.65 A resolution) and actinidin (1.7 A). A detailed comparison of these two structures was performed to determine the effect of amino acid changes on the conformation. It appeared that, despite only 48% identity in their amino acid sequence, different crystallization conditions and different X-ray data collection techniques, their structures are surprisingly similar with a root-mean-square difference of 0.40 A between 76% of the main-chain atoms (differences less than 3 sigma). Insertions and deletions cause larger differences but they alter the conformation over a very limited range of two to three residues only. Conformations of identical side-chains are generally retained to the same extent as the main-chain conformation. If they do change, this is due to a modified local environment. Several examples are described. Spatial positions of hydrogen bonds are conserved to a greater extent than are the specific groups involved. The greatest structural similarity is found for the active site residues of papain and actinidin, for the internal water molecules and for the main-chain conformation of residues in alpha-helices and anti-parallel beta-sheet structure. This was reflected also in the similarity of the temperature factors. It suggests that the secondary structural elements form the skeleton of the molecule and that their interaction is the main factor in directing the fold of the polypeptide chain. Therefore, substitution of residues in the skeleton will, in general, have the most drastic effect on the conformation of the protein molecule. In papain and actinidin, some main-chain-side-chain hydrogen bonds are also strongly conserved and these may determine the folding of non-repetitive parts of the structure. Furthermore, we included primary structure information for three homologous thiol proteases: stem bromelain, and the cathepsins B and H. By combining the three-dimensional structural information for papain and actinidin with sequence homologies and identities, we conclude that the overall folding pattern of the polypeptide chain is grossly the same in all five proteases, and that they utilize the same catalytic mechanism.

Preliminary crystallographic studies on human apo-lactoferrin in its native and deglycosylated forms.

Human apo-lactoferrin in both native and deglycosylated forms has been purified, and crystals obtained by dialysis against low ionic strength buffer solutions. The crystals of native apo-lactoferrin are orthorhombic, space group P2(1)2(1)2(1) with cell dimensions a = 222.0 A, b = 115.6 A, c = 77.8 A and have two protein molecules per asymmetric unit. Two crystal forms of deglycosylated apo-lactoferrin have been obtained. One is orthorhombic, space group P2(1)2(1)2(1), with cell dimensions a = 152.1 A, b = 94.6 A, c = 55.8 A. The second is tetragonal, space group I4, with cell dimensions a = b = 189.4 A, c = 55.1 A. Both of the latter have only one molecule per asymmetric unit, and are suitable for high-resolution X-ray structure analysis.

Structure of azurin from Alcaligenes denitrificans at 2.5 A resolution.

The structure of the blue copper protein, azurin, from Alcaligenes denitrificans has been determined from an electron density map at a nominal resolution of 3.0 A. Four isomorphous heavy-atom derivatives, prepared with KAu(CN)2, uranyl acetate, Hg(NH3)2Cl2 and KAu(CN)2 + uranyl acetate (a double derivative) were used to calculate phases by the method of isomorphous replacement. The overall figure of merit was 0.61. The two molecules in the asymmetric unit are related by an approximate 2-fold axis. Independent interpretations of the density were made for the two molecules, and the structures have since been partially refined. After 12 refinement cycles, using the Hendrickson-Konnert restrained least-squares program, the R factor is 0.318 for data to 2.5 A resolution and there are no major conformational differences between the two molecules. Refinement is continuing. Eight extended strands of the polypeptide chain form a beta-barrel structure whose topology is the same as that of plastocyanin and the alternative folding proposed for Pseudomonas aeruginosa azurin. As in the latter two proteins, the copper atom forms three short bonds, with His-46 N delta 1, His117 N delta 1 and Cys112 S gamma, and one longer bond, with Met121 S delta, these four ligands forming a very distorted tetrahedron. A possible additional interaction, between copper and the carbonyl oxygen of Gly45, cannot be discounted at the present stage of the analysis. A surface hydrophobic patch, around the edge of the imidazole ring of His117 appears the most likely electron transfer locus. The sequences of azurin and plastocyanin have been aligned and the homology between the two proteins is discussed.

Preliminary crystallographic studies of copper(II)- and oxalate-substituted human lactoferrin.

As part of a comparative study on the binding of different metals and anions by human lactoferrin, we have prepared and crystallized: (1) dicupric lactoferrin with Cu2+ and carbonate in each site (Cu2Lf); and (2) a lactoferrin complex with Cu2+ and carbonate in one site, and Cu2+ and oxalate in the other (Cu2oxLf). Crystals of Cu2Lf are orthorhombic: a = 155.9, b = 97.0, c = 56.0 A, space-group P2(1)2(1)2(1); those of Cu2oxLf are also orthorhombici a = 155.9, b = 97.1, c = 56.2 A, space-group P2(1)2(1)2(1). Both are isomorphous with diferric human lactoferrin, Fe2Lf. Diffractometer data to 2.6 A and 2.5 A have been collected for Cu2Lf and Cu2oxLf, respectively. Difference maps show that the main effect of substitution of Cu2+ for Fe3+ is a small shift (0.5 to 1.0 A) in the metal position in each site. For Cu2oxLf the oxalate ion is found to be accommodated in the C-lobe, bound to copper in a bidentate mode, causing only small local changes, in the positions of adjacent Arg and Tyr side-chains.

Structure of human lactoferrin: crystallographic structure analysis and refinement at 2.8 A resolution.

The structure of human lactoferrin has been refined crystallographically at 2.8 A (1 A = 0.1 nm) resolution using restrained least squares methods. The starting model was derived from a 3.2 A map phased by multiple isomorphous replacement with solvent flattening. Rebuilding during refinement made extensive use of these experimental phases, in combination with phases calculated from the partial model. The present model, which includes 681 of the 691 amino acid residues, two Fe3+, and two CO3(2-), gives an R factor of 0.206 for 17,266 observed reflections between 10 and 2.8 A resolution, with a root-mean-square deviation from standard bond lengths of 0.03 A. As a result of the refinement, two single-residue insertions and one 13-residue deletion have been made in the amino acid sequence, and details of the secondary structure and tertiary interactions have been clarified. The two lobes of the molecule, representing the N-terminal and C-terminal halves, have very similar folding, with a root-mean-square deviation, after superposition, of 1.32 A for 285 out of 330 C alpha atoms; the only major differences being in surface loops. Each lobe is subdivided into two dissimilar alpha/beta domains, one based on a six-stranded mixed beta-sheet, the other on a five-stranded mixed beta-sheet, with the iron site in the interdomain cleft. The two iron sites appear identical at the present resolution. Each iron atom is coordinated to four protein ligands, 2 Tyr, 1 Asp, 1 His, and the specific Co3(2-), which appears to bind to iron in a bidentate mode. The anion occupies a pocket between the iron and two positively charged groups on the protein, an arginine side-chain and the N terminus of helix 5, and may serve to neutralize this positive charge prior to iron binding. A large internal cavity, beyond the Arg side-chain, may account for the binding of larger anions as substitutes for CO3(2-). Residues on the other side of the iron site, near the interdomain crossover strands could provide secondary anion binding sites, and may explain the greater acid-stability of iron binding by lactoferrin, compared with serum transferrin. Interdomain and interlobe interactions, the roles of charged side-chains, heavy-atom binding sites, and the construction of the metal site in relation to the binding of different metals are also discussed.

Three-dimensional structure of diferric bovine lactoferrin at 2.8 A resolution.

The three-dimensional structure of diferric bovine lactoferrin (bLf) has been determined by X-ray crystallography in order to investigate the factors that influence iron binding and release by transferrins. The structure was solved by molecular replacement, using the coordinates of diferric human lactoferrin (hLf) as a search model, and was refined with data to 2.8 A resolution by simulated annealing (X-PLOR) and restrained least squares (TNT). The final model comprises 5310 protein atoms (residues 5 to 689), 124 carbohydrate atoms (from ten monosaccharide units, in three glycan chains), 2 Fe3+, 2 CO32- and 50 water molecules. This model gives an R-factor of 0.232 for 21440 reflections in the resolution range 30.0 to 2.8 A. The folding of the bLf molecule is essentially the same as that of hLf, but bLf differs in the extent of closure of the two domains of each lobe, and in the relative orientations of the two lobes. Differences in domain closure are attributed to amino acid changes in the interface, and differences in lobe orientations to slightly altered packing of two hydrophobic patches between the lobes. Changed interdomain interactions may explain the lesser iron affinity of bLf, compared with hLf, and two lysine residues behind the N-lobe iron site of bLf offer new insights into the "dilysine trigger" mechanism proposed for iron release by transferrins. The bLf structure is also notable for several well-defined oligosaccharide units which demonstrate the structural factors that stabilise carbohydrate structure. One glycan chain, attached to Asn545, appears to contribute to interdomain interactions and may modulate iron release from the C-lobe.