Molecular pathology of X linked retinoschisis: mutations interfere with retinoschisin secretion and oligomerisation.

BACKGROUND/AIM: X linked retinoschisis (XLRS) is caused by mutations in RS1 which encodes the discoidin domain protein retinoschisin, secreted by photoreceptors and bipolar cells. Missense mutations occur throughout the gene and some of these are known to interfere with protein secretion. This study was designed to investigate the functional consequences of missense mutations at different locations in retinoschisin. METHODS AND RESULTS: The authors developed a structural model of the retinoschisin discoidin domain and used this to predict the effects of missense mutations. They expressed disease associated mutations and found that those affecting conserved residues prevented retinoschisin secretion. Most of the remaining mutations cluster within a series of loops on the surface of the beta barrel structure and do not interfere with secretion, suggesting this region may be a ligand binding site. They also demonstrated that wild type retinoschisin octamerises and associates with the cell surface. A subgroup of secreted mutations reduce oligomerisation (C59S, C219G, C223R). CONCLUSIONS: It is suggested that there are three different molecular mechanisms which lead to XLRS: mutations interfering with secretion, mutations interfering with oligomerisation, and mutations that allow secretion and oligomerisation but interfere with retinoschisin function. The authors conclude that binding of oligomerised retinoschisin at the cell surface is important in its presumed role in cell adhesion.

Pushing the boundaries of molecular replacement with maximum likelihood.

The molecular-replacement method works well with good models and simple unit cells, but often fails with more difficult problems. Experience with likelihood in other areas of crystallography suggests that it would improve performance significantly. For molecular replacement, the form of the required likelihood function depends on whether there is ambiguity in the relative phases of the contributions from symmetry-related molecules (e.g. rotation versus translation searches). Likelihood functions used in structure refinement are appropriate only for translation (or six-dimensional) searches, where the correct translation will place all of the atoms in the model approximately correctly. A new likelihood function that allows for unknown relative phases is suitable for rotation searches. It is shown that correlations between sequence identity and coordinate error can be used to calibrate parameters for model quality in the likelihood functions. Multiple models of a molecule can be combined in a statistically valid way by setting up the joint probability distribution of the true and model structure factors as a multivariate complex normal distribution, from which the conditional distribution of the true structure factor given the models can be derived. Tests in a new molecular-replacement program, Beast, show that the likelihood-based targets are more sensitive and more accurate than previous targets. The new multiple-model likelihood function has a dramatic impact on success.

Different structural requirements for plasminogen activator inhibitor 1 (PAI-1) during latency transition and proteinase inhibition as evidenced by phage-displayed hypermutated PAI-1 libraries.

Plasminogen activator inhibitor type 1 (PAI-1) is a member of the serine protease inhibitor (serpin) superfamily. Its highly mobile reactive-center loop (RCL) is thought to account for both the rapid inhibition of tissue-type plasminogen activator (t-PA), and the rapid and spontaneous transition of the unstable, active form of PAI-1 into a stable, inactive (latent) conformation (t(1/2) at 37 degrees C, 2.2 hours). We determined the amino acid residues responsible for the inherent instability of PAI-1, to assess whether these properties are independent and, consequently, whether the structural basis for inhibition and latency transition is different. For that purpose, a hypermutated PAI-1 library that is displayed on phage was pre-incubated for increasing periods (20 to 72 hours) at 37 degrees C, prior to a stringent selection for rapid t-PA binding. Accordingly, four rounds of phage-display selection resulted in the isolation of a stable PAI-1 variant (st-44: t(1/2) 450 hours) with 11 amino acid mutations. Backcrossing by DNA shuffling of this stable mutant with wt PAI-1 was performed to eliminate non-contributing mutations. It was shown that the combination of mutations at positions 50, 56, 61, 70, 94, 150, 222, 223, 264 and 331 increases the half-life of PAI-1 245-fold. Furthermore, within the limits of detection the stable mutants isolated are functionally indistinguishable from wild-type PAI-1 with respect to the rate of inhibition of t-PA, cleavage by t-PA, and binding to vitronectin. These stabilizing mutations constitute largely reversions to the stable "serpin consensus sequence" and are located in areas implicated in PAI-1 stability (e.g. the vitronectin-binding domain and the proximal hinge). Collectively, our data provide evidence that the structural requirements for PAI-1 loop insertion during latency transition and target proteinase inhibition can be separated.

Immunoprophylactic potential of cloned Shiga toxin 2 B subunit.

The Shiga toxins Stx1 and Stx2 contribute to the development of enterohemorrhagic O157:H7 Escherichia coli-mediated colitis and hemolytic-uremic syndrome in humans. The Stx2 B subunit, which binds to globotriaosylceramide (GB3) receptors on target cells, was cloned. This involved replacing the Stx2 B subunit leader peptide nucleotide sequences with those from the Stx1 B subunit. The construct was expressed in the TOPP3 E. coli strain. The Stx2 B subunits from this strain assembled into a pentamer and bound to a GB3 receptor analogue. The cloned Stx2 B subunit was not cytotoxic to Vero cells or apoptogenic in Burkitt's lymphoma cells. Although their immune response to the Stx2 B subunit was variable, rabbits that developed Stx2 B subunit-specific antibodies, as determined by immunoblot and in vitro cytotoxicity neutralization assays, survived a challenge with Stx2 holotoxin. This is thought to be the first demonstration of the immunoprophylactic potential of the Stx2 B subunit.

Structure of a serpin-protease complex shows inhibition by deformation.

The serpins have evolved to be the predominant family of serine-protease inhibitors in man. Their unique mechanism of inhibition involves a profound change in conformation, although the nature and significance of this change has been controversial. Here we report the crystallographic structure of a typical serpin-protease complex and show the mechanism of inhibition. The conformational change is initiated by reaction of the active serine of the protease with the reactive centre of the serpin. This cleaves the reactive centre, which then moves 71 A to the opposite pole of the serpin, taking the tethered protease with it. The tight linkage of the two molecules and resulting overlap of their structures does not affect the hyperstable serpin, but causes a surprising 37% loss of structure in the protease. This is induced by the plucking of the serine from its active site, together with breakage of interactions formed during zymogen activation. The disruption of the catalytic site prevents the release of the protease from the complex, and the structural disorder allows its proteolytic destruction. It is this ability of the conformational mechanism to crush as well as inhibit proteases that provides the serpins with their selective advantage.

Crystal structure of Pseudomonas aeruginosa PAK pilin suggests a main-chain-dominated mode of receptor binding.

Fibers of pilin monomers (pili) form the dominant adhesin of Pseudomonas aeruginosa, and they play an important role in infections by this opportunistic bacterial pathogen. Blocking adhesion is therefore a target for vaccine development. The receptor-binding site is located in a C-terminal disulphide-bonded loop of each pilin monomer, but functional binding sites are displayed only at the tip of the pilus. A factor complicating vaccination is that different bacterial strains produce distinct, and sometimes highly divergent, pilin variants. It is surprising that all strains still appear to bind a common receptor, asialo-GM1. Here, we present the 1.63 A crystal structure of pilin from P. aeruginosa strain PAK. The structure shows that the proposed receptor-binding site is formed by two beta-turns that create a surface dominated by main-chain atoms. Receptor specificity could therefore be maintained, whilst allowing side-chain variation, if the main-chain conformation is conserved. The location of the binding site relative to the proposed packing of the pilus fiber raises new issues and suggests that the current fiber model may have to be reconsidered. Finally, the structure of the C-terminal disulphide-bonded loop will provide the template for the structure-based design of a consensus sequence vaccine.

A mutant Shiga-like toxin IIe bound to its receptor Gb(3): structure of a group II Shiga-like toxin with altered binding specificity.

BACKGROUND: Shiga-like toxins (SLTs) are produced by the pathogenic strains of Escherichia coli that cause hemorrhagic colitis and hemolytic uremic syndrome. These diseases in humans are generally associated with group II family members (SLT-II and SLT-IIc), whereas SLT-IIe (pig edema toxin) is central to edema disease of swine. The pentameric B-subunit component of the majority of family members binds to the cell-surface glycolipid globotriaosyl ceramide (Gb(3)), but globotetraosyl ceramide (Gb(4)) is the preferred receptor for SLT-IIe. A double-mutant of the SLT-IIe B subunit that reverses two sequence differences from SLT-II (GT3; Gln65-->Glu, Lys67-->Gln, SLT-I numbering) has been shown to bind more strongly to Gb(3) than to Gb(4). RESULTS: To understand the molecular basis of receptor binding and specificity, we have determined the structure of the GT3 mutant B pentamer, both in complex with a Gb(3) analogue (2.0 A resolution; R = 0.155, R(free) = 0.194) and in its native form (2.35 A resolution; R = 0.187, R(free) = 0.232). CONCLUSIONS: These are the first structures of a member of the medically important group II Shiga-like toxins to be reported. The structures confirm the previous observation of multiple binding sites on each SLT monomer, although binding site 3 is not occupied in the GT3 structure. Analysis of the binding properties of mutants suggests that site 3 is a secondary Gb(4)-binding site. The two mutated residues are located appropriately to interact with the extra betaGalNAc residue on Gb(4). Differences in the binding sites provide a molecular basis for understanding the tissue specificities and pathogenic mechanisms of members of the SLT family.

Shiga-like toxins are neutralized by tailored multivalent carbohydrate ligands.

The diseases caused by Shiga and cholera toxins account for the loss of millions of lives each year. Both belong to the clinically significant subset of bacterial AB5 toxins consisting of an enzymatically active A subunit that gains entry to susceptible mammalian cells after oligosaccharide recognition by the B5 homopentamer. Therapies might target the obligatory oligosaccharide-toxin recognition event, but the low intrinsic affinity of carbohydrate-protein interactions hampers the development of low-molecular-weight inhibitors. The toxins circumvent low affinity by binding simultaneously to five or more cell-surface carbohydrates. Here we demonstrate the use of the crystal structure of the B5 subunit of Escherichia coli O157:H7 Shiga-like toxin I (SLT-I) in complex with an analogue of its carbohydrate receptor to design an oligovalent, water-soluble carbohydrate ligand (named STARFISH), with subnanomolar inhibitory activity. The in vitro inhibitory activity is 1-10-million-fold higher than that of univalent ligands and is by far the highest molar activity of any inhibitor yet reported for Shiga-like toxins I and II. Crystallography of the STARFISH/Shiga-like toxin I complex explains this activity. Two trisaccharide receptors at the tips of each of five spacer arms simultaneously engage all five B subunits of two toxin molecules.

Inactive conformation of the serpin alpha(1)-antichymotrypsin indicates two-stage insertion of the reactive loop: implications for inhibitory function and conformational disease.

The serpins are a family of proteinase inhibitors that play a central role in the control of proteolytic cascades. Their inhibitory mechanism depends on the intramolecular insertion of the reactive loop into beta-sheet A after cleavage by the target proteinase. Point mutations within the protein can allow aberrant conformational transitions characterized by beta-strand exchange between the reactive loop of one molecule and beta-sheet A of another. These loop-sheet polymers result in diseases as varied as cirrhosis, emphysema, angio-oedema, and thrombosis, and we recently have shown that they underlie an early-onset dementia. We report here the biochemical characteristics and crystal structure of a naturally occurring variant (Leu-55-Pro) of the plasma serpin alpha(1)-antichymotrypsin trapped as an inactive intermediate. The structure demonstrates a serpin configuration with partial insertion of the reactive loop into beta-sheet A. The lower part of the sheet is filled by the last turn of F-helix and the loop that links it to s3A. This conformation matches that of proposed intermediates on the pathway to complex and polymer formation in the serpins. In particular, this intermediate, along with the latent and polymerized conformations, explains the loss of activity of plasma alpha(1)-antichymotrypsin associated with chronic obstructive pulmonary disease in patients with the Leu-55-Pro mutation.

A 2.6 A structure of a serpin polymer and implications for conformational disease.

The function of the serpins as proteinase inhibitors depends on their ability to insert the cleaved reactive centre loop as the fourth strand in the main A beta-sheet of the molecule upon proteolytic attack at the reactive centre, P1-P1'. This mechanism is vulnerable to mutations which result in inappropriate intra- or intermolecular loop insertion in the absence of cleavage. Intermolecular loop insertion is known as serpin polymerisation and results in a variety of diseases, most notably liver cirrhosis resulting from mutations of the prototypical serpin alpha1-antitrypsin. We present here the 2.6 A structure of a polymer of alpha1-antitrypsin cleaved six residues N-terminal to the reactive centre, P7-P6 (Phe352-Leu353). After self insertion of P14 to P7, intermolecular linkage is affected by insertion of the P6-P3 residues of one molecule into the partially occupied beta-sheet A of another. This results in an infinite, linear polymer which propagates in the crystal along a 2-fold screw axis. These findings provide a framework for understanding the uncleaved alpha1-antitrypsin polymer and fibrillar and amyloid deposition of proteins seen in other conformational diseases, with the ordered array of polymers in the crystal resulting from slow accretion of the cleaved serpin over the period of a year.

Detecting outliers in non-redundant diffraction data.

Outliers are observations which are very unlikely to be correct, as judged by independent observations or other prior information. Such unexpected observations are treated, effectively, as being more informative about possible models, so they can seriously impede the course of structure determination and refinement. The best way to detect and eliminate outliers is to collect highly redundant data, but it is not always possible to make multiple measurements of every reflection. For non-redundant data, the prior expectation given either by a Wilson distribution of intensities or model-based structure-factor probability distributions can be used to detect outliers. This captures mostly the excessively strong reflections, which dominate the features of electron-density maps or, even more so, Patterson maps. The outlier rejection tests have been implemented in a program, Outliar.

The active conformation of plasminogen activator inhibitor 1, a target for drugs to control fibrinolysis and cell adhesion.

BACKGROUND: Plasminogen activator inhibitor 1 (PAI-1) is a serpin that has a key role in the control of fibrinolysis through proteinase inhibition. PAI-1 also has a role in regulating cell adhesion processes relevant to tissue remodeling and metastasis; this role is mediated by its binding to the adhesive glycoprotein vitronectin rather than by proteinase inhibition. Active PAI-1 is metastable and spontaneously transforms to an inactive latent conformation. Previous attempts to crystallize the active conformation of PAI-1 have failed. RESULTS: The crystal structure of a stable quadruple mutant of PAI-1(Asn150-->His, Lys154-->Thr, Gln319-->Leu, Met354-->Ile) in its active conformation has been solved at a nominal 3 A resolution. In two of four independent molecules within the crystal, the flexible reactive center loop is unconstrained by crystal-packing contacts and is disordered. In the other two molecules, the reactive center loop forms intimate loop-sheet interactions with neighboring molecules, generating an infinite chain within the crystal. The overall conformation resembles that seen for other active inhibitory serpins. CONCLUSIONS: The structure clarifies the molecular basis of the stabilizing mutations and the reduced affinity of PAI-1, on cleavage or in the latent form, for vitronectin. The infinite chain of linked molecules also suggests a new mechanism for the serpin polymerization associated with certain diseases. The results support the concept that the reactive center loop of an active serpin is flexible and has no defined conformation in the absence of intermolecular contacts. The determination of the structure of the active form constitutes an essential step for the rational design of PAI-1 inhibitors.

The identification of three biologically relevant globotriaosyl ceramide receptor binding sites on the Verotoxin 1 B subunit.

The Verotoxin 1 (VT1) B subunit binds to the glycosphingolipid receptor globotriaosylceramide (Gb3). Receptor-binding specificity is associated with the terminally linked Galalpha(1-4) Galbeta disaccharide sequence of the receptor. Recently, three globotriose (Galalpha[1-4] Galbeta [1-4] Glcbeta) binding sites per B-subunit monomer were identified by crystallography. Two of these sites (sites I and II) are located adjacent to phenylalanine-30. Site I was originally predicted as a potential Gb3 binding site on the basis of sequence conservation, and site II was additionally predicted based on computer modelling and receptor docking. The third (site III) was also identified by crystallography and is located at the N-terminal end of the alpha-helix. To determine the biological significance of sites II and III, and to support our previous findings of the significance of site I, we examined the binding properties and cytotoxicity of VT1 mutants designed to block Gb3 binding at each site selectively. The Scatchard analysis of saturation-binding data for each mutant revealed that only the amino acid substitutions predicted to affect site I (D-17E) or site II (G-62T) caused reductions in the binding affinity and capacity of VT1 for Gb3. Similarly, those mutations at sites I and II also caused significant reductions in both Vero and MRC-5 cell cytotoxicity (by seven and five logs, respectively, for G-62T and by four and two logs, respectively, for D-17E). In contrast, the substitution of alanine for W-34 at site III did not reduce the high-affinity binding of the B subunit, despite causing a fourfold reduction in the receptor-binding capacity. The corresponding mutant W-34A holotoxin had a two-log reduction in cytotoxicity on Vero cells and no statistically significant reduction on MRC-5 cells. We conclude that the high-affinity receptor binding most relevant for cell cytotoxicity occurs at sites I and II. In contrast, site III appears to mediate the recognition of additional Gb3 receptor epitopes but with lower affinity. Our results support the significance of the indole ring of W-34 for binding at this site.

Extending the limits of molecular replacement through combined simulated annealing and maximum-likelihood refinement.

Phases determined by the molecular-replacement method often suffer from model bias. In extreme cases, the refinement of the atomic model can stall at high free R values when the resulting electron-density maps provide little indication of how to correct the model, sometimes rendering even a correct solution unusable. Here, it is shown that several recent advances in refinement methodology allow productive refinement, even in cases where the molecular-replacement-phased electron-density maps do not allow manual rebuilding. In test calculations performed with a series of homologous models of penicillopepsin using either backbone atoms, or backbone atoms plus conserved core residues, model bias is reduced and refinement can proceed efficiently, even if the initial model is far from the correct one. These new methods combine cross-validation, torsion-angle dynamics simulated annealing and maximum-likelihood target functions. It is also shown that the free R value is an excellent indicator of model quality after refinement, potentially discriminating between correct and incorrect molecular-replacement solutions. The use of phase information, even in the form of bimodal single-isomorphous-replacement phase distributions, greatly improves the radius of convergence of refinement and hence the quality of the electron-density maps, further extending the limits of molecular replacement.

Crystallography & NMR system: A new software suite for macromolecular structure determination.

A new software suite, called Crystallography & NMR System (CNS), has been developed for macromolecular structure determination by X-ray crystallography or solution nuclear magnetic resonance (NMR) spectroscopy. In contrast to existing structure-determination programs, the architecture of CNS is highly flexible, allowing for extension to other structure-determination methods, such as electron microscopy and solid-state NMR spectroscopy. CNS has a hierarchical structure: a high-level hypertext markup language (HTML) user interface, task-oriented user input files, module files, a symbolic structure-determination language (CNS language), and low-level source code. Each layer is accessible to the user. The novice user may just use the HTML interface, while the more advanced user may use any of the other layers. The source code will be distributed, thus source-code modification is possible. The CNS language is sufficiently powerful and flexible that many new algorithms can be easily implemented in the CNS language without changes to the source code. The CNS language allows the user to perform operations on data structures, such as structure factors, electron-density maps, and atomic properties. The power of the CNS language has been demonstrated by the implementation of a comprehensive set of crystallographic procedures for phasing, density modification and refinement. User-friendly task-oriented input files are available for nearly all aspects of macromolecular structure determination by X-ray crystallography and solution NMR.

Fragment-based modeling of NAD binding to the catalytic subunits of diphtheria and pertussis toxins.

We describe a novel application of a fragment-based ligand docking technique; similar methods are commonly applied to the de novo design of ligands for target protein binding sites. We have used several new flexible docking and superposition tools, as well as a more conventional rigid-body (fragment) docking method, to examine NAD binding to the catalytic subunits of diphtheria (DT) and pertussis (PT) toxins, and to propose a model of the NAD-PT complex. Docking simulations with the rigid NAD fragments adenine and nicotinamide revealed that the low-energy dockings clustered in three distinct sites on the two proteins. Two of the sites were common to both fragments and were related to the structure of NAD bound to DT in an obvious way; however, the adenine subsite of PT was shifted relative to that of DT. We chose adenine/nicotinamide pairs of PT dockings from these clusters and flexibly superimposed NAD onto these pairs. A Monte Carlo-based flexible docking procedure and energy minimization were used to refine the modeled NAD-PT complexes. The modeled complex accounts for the sequence and structural similarities between PT and DT and is consistent with many results that suggest the catalytic importance of certain residues. A possible functional role for the structural difference between the two complexes is discussed.

Structure of the shiga-like toxin I B-pentamer complexed with an analogue of its receptor Gb3.

Shiga-like toxin I (SLT-I) is a virulence factor of Escherichia coli strains that cause disease in humans. Like other members of the Shiga toxin family, it consists of an enzymatic (A) subunit and five copies of a binding subunit (the B-pentamer). The B-pentamer binds to a specific glycolipid, globotriaosylceramide (Gb3), on the surface of target cells and thereby plays a crucial role in the entry of the toxin. Here we present the crystal structure at 2.8 A resolution of the SLT-I B-pentamer complexed with an analogue of the Gb3 trisaccharide. The structure reveals a surprising density of binding sites, with three trisaccharide molecules bound to each B-subunit monomer of 69 residues. All 15 trisaccharides bind to one side of the B-pentamer, providing further evidence that this side faces the cell membrane. The structural model is consistent with data from site-directed mutagenesis and binding of carbohydrate analogues, and allows the rational design of therapeutic Gb3 analogues that block the attachment of toxin to cells.

Modeling the carbohydrate-binding specificity of pig edema toxin.

The wild-type binding pentamer of Shiga-like toxin IIe (SLT-IIe) binds both the globotriaosylceramide (Gb3) and globotetraosylceramide (Gb4) cell surface glycolipids, whereas the double mutant GT3 (Q65E/K67Q) exhibits a marked preference for Gb3 [Tyrrell, G. J., et al. (1992) Proc. Natl. Acad. Sci. U.S.A. 89, 524-528]. We modeled three unique sites (sites 1-3) for binding of the carbohydrate moiety of Gb3 to GT3 and SLT-IIe, on the basis of the three sites observed for the SLT-I pentamer [Ling, H., et al. (1998) Biochemistry 37, 1777-1788]. Examination of the three sites in light of various mutation and binding data strongly suggested that one of the binding sites plays a role in the change of specificity observed for the GT3 mutant. We applied several modeling techniques, and developed a model for binding of the carbohydrate moiety of Gb4 to this site of the SLT-IIe binding pentamer. This model is consistent with a wide variety of mutation and binding data and clearly shows the importance of the terminal GalNAc residue of Gb4, as well as that of the two mutated residues of GT3, to the intermolecular interaction.

Incorporation of prior phase information strengthens maximum-likelihood structure refinement.

The application of a maximum-likelihood analysis to the problem of structure refinement has led to striking improvements over the traditional least-squares methods. Since the method of maximum likelihood allows for a rational incorporation of other sources of information, we have derived a likelihood function that incorporates experimentally determined phase information. In a number of different test cases, this target function performs better than either a least-squares target or a maximum-likelihood function lacking prior phases. Furthermore, this target gives significantly better results compared with other functions incorporating phase information. When combined with a procedure to mask 'unexplained' density, the phased likelihood target also makes it possible to refine very incomplete models.

Accumulating evidence suggests that several AB-toxins subvert the endoplasmic reticulum-associated protein degradation pathway to enter target cells.

Several AB-toxins appear to have independently evolved mechanisms by which they undergo retrograde transport from the cell membrane to the endoplasmic reticulum (ER). Recent insights into ER-associated protein degradation (ERAD) now provide clues as to why these toxins have selected the ER as the site of cell entry. We propose that they disguise themselves as misfolded proteins to enter the ERAD pathway. We further link the observation that these toxins have few, if any, lysine residues to the need to escape ubiquitin-mediated protein degradation, the ultimate destination of the ERAD pathway. The actual membrane translocation step remains unclear, but studies on viral immune evasion mechanisms indicate that retrotranslocation across the ER lipid bilayer may involve SEC61. Understanding the internalization process of these toxins opens new avenues for preventing their entry into cells. In addition, this knowledge can be exploited to create protein-based pharmaceuticals that act on cytosolic targets.