Amyloid fibers are water-filled nanotubes.

A study of papers on amyloid fibers suggested to us that cylindrical beta-sheets are the only structures consistent with some of the x-ray and electron microscope data. We then found that our own 7-year-old and hitherto enigmatic x-ray diagram of poly-L-glutamine fits a cylindrical sheet of 31 A diameter made of beta-strands with 20 residues per helical turn. Successive turns are linked by hydrogen bonds between both the main chain and side chain amides, and side chains point alternately into and out of the cylinder. Fibers of the exon-1 peptide of huntingtin and of the glutamine- and asparagine-rich region of the yeast prion Sup35 give the same underlying x-ray diagrams, which show that they have the same structure. Electron micrographs show that the 100-A-thick fibers of the Sup35 peptide are ropes made of three protofibrils a little over 30 A thick. They have a measured mass of 1,450 Da/A, compared with 1,426 Da/A for a calculated mass of three protofibrils each with 20 residues per helical turn wound around each other with a helical pitch of 510 A. Published x-ray diagrams and electron micrographs show that fibers of synuclein, the protein that forms the aggregates of Parkinson disease, consist of single cylindrical beta-sheets. Fibers of Alzheimer A beta fragments and variants are probably made of either two or three concentric cylindrical beta-sheets. Our structure of poly-L-glutamine fibers may explain why, in all but one of the neurodegenerative diseases resulting from extension of glutamine repeats, disease occurs when the number of repeats exceeds 37-40. A single helical turn with 20 residues would be unstable, because there is nothing to hold it in place, but two turns with 40 residues are stabilized by the hydrogen bonds between their amides and can act as nuclei for further helical growth. The A beta peptide of Alzheimer's disease contains 42 residues, the best number for nucleating further growth. All these structures are very stable; the best hope for therapies lies in preventing their growth.

Modularity and homology: modelling of the titin type I modules and their interfaces.

Titin is a giant muscle protein with a highly modular architecture consisting of multiple repeats of two sequence motifs, named type I and type II. Type I motifs are homologous to members of the fibronectin type 3 (Fn3) superfamily, one of the motifs most widespread in modular proteins. Fn3 domains are thought to mediate protein-protein interactions and to act as spacers. In titin, Fn3 modules are present in two different super-repeated patterns, likely to be involved in sarcomere assembly through interactions with A-band proteins. Here, we discuss results from homology modelling the whole family of Fn3 domains in titin. Homology modelling is a powerful tool that will play an increasingly important role in the post-genomic era. It is particularly useful for extending experimental structure determinations of parts of multidomain proteins that contain multiple copies of the same motif. The 3D structures of a representative titin type I domain and of other extracellular Fn3 modules were used as a template to model the structures of the 132 copies in titin. The resulting models suggest residues that contribute to the fold stability and allow us to distinguish these from residues likely to have functional importance. In particular, analysis of the models and mapping of the consensus sequence onto the 3D structure suggest putative surfaces of interaction with other proteins. From the structures of isolated modules and the pattern of conservation in the multiple alignment of the whole titin Ig and Fn3 families, it is possible to address the question of how tandem modules are assembled. Our predictions can be validated experimentally.

Protein structural alignments and functional genomics.

Structural genomics-the systematic solution of structures of the proteins of an organism-will increasingly often produce molecules of unknown function with no close relative of known function. Prediction of protein function from structure has thereby become a challenging problem of computational molecular biology. The strong conservation of active site conformations in homologous proteins suggests a method for identifying them. This depends on the relationship between size and goodness-of-fit of aligned substructures in homologous proteins. For all pairs of proteins studied, the root-mean-square deviation (RMSD) as a function of the number of residues aligned varies exponentially for large common substructures and linearly for small common substructures. The exponent of the dependence at large common substructures is well correlated with the RMSD of the core as originally calculated by Chothia and Lesk (EMBO J 1986;5:823-826), affording the possibility of reconciling different structural alignment procedures. In the region of small common substructures, reduced aligned subsets define active sites and can be used to suggest the locations of active sites in homologous proteins.

Assessment of novel fold targets in CASP4: predictions of three-dimensional structures, secondary structures, and interresidue contacts.

In the Novel Fold category, three types of predictions were assessed: three-dimensional structures, secondary structures, and residue-residue contacts. For predictions of three-dimensional models, CASP4 targets included 5 domains or structures with novel folds, and 13 on the borderline between Novel Fold and Fold Recognition categories. These elicited 1863 predictions of these and other targets by methods more general than comparative modeling or fold recognition techniques. The group of Bonneau, Tsai, Ruczinski, and Baker stood out as performing well with the greatest consistency. In many cases, several groups were able to predict fragments of the target correctly-often at a level somewhat larger than standard supersecondary structures-but were not able to assemble fragments into a correct global topology. The methods of Bonneau, Tsai, Ruczinski, and Baker have been successful in addressing the fragment assembly problem for many but not all the target structures.

Phylogeny of the serpin superfamily: implications of patterns of amino acid conservation for structure and function.

We present a comprehensive alignment and phylogenetic analysis of the serpins, a superfamily of proteins with known members in higher animals, nematodes, insects, plants, and viruses. We analyze, compare, and classify 219 proteins representative of eight major and eight minor subfamilies, using a novel technique of consensus analysis. Patterns of sequence conservation characterize the family as a whole, with a clear relationship to the mechanism of function. Variations of these patterns within phylogenetically distinct groups can be correlated with the divergence of structure and function. The goals of this work are to provide a carefully curated alignment of serpin sequences, to describe patterns of conservation and divergence, and to derive a phylogenetic tree expressing the relationships among the members of this family. We extend earlier studies by Huber and Carrell as well as by Marshall, after whose publication the serpin family has grown functionally, taxonomically, and structurally. We used gene and protein sequence data, crystal structures, and chromosomal location where available. The results illuminate structure-function relationships in serpins, suggesting roles for conserved residues in the mechanism of conformational change. The phylogeny provides a rational evolutionary framework to classify serpins and enables identification of conserved amino acids. Patterns of conservation also provide an initial point of comparison for genes identified by the various genome projects. New homologs emerging from sequencing projects can either take their place within the current classification or, if necessary, extend it.

Conformational changes in serpins: II. The mechanism of activation of antithrombin by heparin.

Antithrombin, uniquely among plasma serpins acting as proteinase inhibitors in the control of the blood coagulation cascade, circulates in a relatively inactive form. Its activation by heparin, and specifically by a pentasaccharide core of heparin, has been shown to involve release of the peptide loop containing the reactive centre from partial insertion in the A sheet of the molecule. Here we compare the structures of the circulating inactive form of antithrombin with the activated structure in complex with heparin pentasaccharide. We show that the rearrangement of the reactive centre loop that occurs upon activation is part of a widespread conformational change involving a realignment of the two major domains of the molecule. We also examine natural mutants that possess high affinity for heparin pentasaccharide, and relate the kinetics of their interaction with heparin pentasaccharide to the structural transitions occuring in the activation process.

Antibody modeling: implications for engineering and design.

Our understanding of the rules relating sequence to structure in antibodies has led to the development of accurate knowledge-based procedures for antibody modeling. Information gained from the analysis of antibody structures has been successfully exploited to engineer antibody-like molecules endowed with prescribed properties, such as increased stability or different specificity, many of which have a broad spectrum of applications both in therapy and in research. Here we describe a knowledge-based procedure for the prediction of the antibody-variable domains, based on the canonical structures method for the antigen-binding site, and discuss its expected accuracy and limitations. The rational design of antibody-based molecules is illustrated using as an example one of the most widely employed modifications of antibody structures: the humanization of animal-derived antibodies to reduce their immunogenicity for serotherapy in humans.

Conformational changes in serpins: I. The native and cleaved conformations of alpha(1)-antitrypsin.

The serpins (SERine Proteinase INhibitors) are a family of proteins with important physiological roles, including but not limited to the inhibition of chymotrypsin-like serine proteinases. The inhibitory mechan- ism involves a large conformational change known as the S-->R (stressed-->relaxed) transition. The largest structural differences occur in a region around the scissile bond called the reactive centre loop: In the native (S) state, the reactive centre is exposed, and is free to interact with proteinases. In inhibitory serpins, in the cleaved (R) state the reactive centre loop forms an additional strand within the beta-sheet. The latent state is an uncleaved state in which the intact reactive centre loop is integrated into the A sheet as in the cleaved form, to give an alternative R state. The serpin structures illustrate detailed control of conformation within a single protein. Serpins are also an unusual family of proteins in which homologues have native states with different folding topologies. Determination of the structures of inhibitory serpins in multiple conformational states permits a detailed analysis of the mechanism of the S-->R transition, and of the way in which a single sequence can form two stabilised states of different topology. Here we compare the conformations of alpha(1)-antitrypsin in native and cleaved states. Many protein conformational changes involve relative motions of large rigid subunits. We determine the rigid subunits of alpha(1)-antitrypsin and analyse the changes in their relative position and orientation. Knowing that the conformational change is initiated by cleavage at the reactive centre, we describe a mechanism of the S-->R transition as a logical sequence of mechanical effects, even though the transition likely proceeds in a concerted manner.

Canonical structures for the hypervariable regions of T cell alphabeta receptors.

T cell alphabeta receptors have binding sites for peptide-MHC complexes formed by six hypervariable regions. Analysis of the six atomic structures known for Valpha and for Vbeta domains shows that their first and second hypervariable regions have one of three or four different main-chain conformations (canonical structures). Six of these canonical structures have the same conformation in complexes with peptide-MHC complexes, the free receptor and/or in an isolated V domain. Thus, for at least the first and second hypervariable regions in the currently known structures, the conformation of the canonical structures is well defined in the free state and is conserved on formation of complexes with peptide-MHC. We identified the key residues that are mainly responsible for the conformation of each canonical structure. The first and second hypervariable regions of Valpha and Vbeta domains are encoded by the germline V segments. Humans have 37 functional Valpha segments and 47 Vbeta segments, and mice have 20 Vbeta segments. Inspection of the size of their hypervariable regions, and of sites that contain key residues, indicates that close to 70 % of Valpha segments and 90 % of Vbeta segments have hypervariable regions with a conformation of one of the known canonical structures. The alpha and beta V gene segments in both humans and mice have only a few combinations of different canonical structure in their first and second hypervariable regions. In human Vbeta domains, the number of different sequences with these canonical structure combinations is larger than in mice, whilst for Valpha domains it is probably smaller.

Conformational changes in serpins: I. The native and cleaved conformations of alpha(1)-antitrypsin.

The serpins (SERine Proteinase INhibitors) are a family of proteins with important physiological roles, including but not limited to the inhibition of chymotrypsin-like serine proteinases. The inhibitory mechan- ism involves a large conformational change known as the S-->R (stressed-->relaxed) transition. The largest structural differences occur in a region around the scissile bond called the reactive centre loop: In the native (S) state, the reactive centre is exposed, and is free to interact with proteinases. In inhibitory serpins, in the cleaved (R) state the reactive centre loop forms an additional strand within the beta-sheet. The latent state is an uncleaved state in which the intact reactive centre loop is integrated into the A sheet as in the cleaved form, to give an alternative R state. The serpin structures illustrate detailed control of conformation within a single protein. Serpins are also an unusual family of proteins in which homologues have native states with different folding topologies. Determination of the structures of inhibitory serpins in multiple conformational states permits a detailed analysis of the mechanism of the S-->R transition, and of the way in which a single sequence can form two stabilised states of different topology. Here we compare the conformations of alpha(1)-antitrypsin in native and cleaved states. Many protein conformational changes involve relative motions of large rigid subunits. We determine the rigid subunits of alpha(1)-antitrypsin and analyse the changes in their relative position and orientation. Knowing that the conformational change is initiated by cleavage at the reactive centre, we describe a mechanism of the S-->R transition as a logical sequence of mechanical effects, even though the transition likely proceeds in a concerted manner.

Tendamistat surface accessibility to the TEMPOL paramagnetic probe.

TEMPOL, the soluble spin-label 4-hydroxy-2,2,6,6-tetramethyl-piperidine-1-oxyl, has been used to determine the surface characteristics of tendamistat, a small protein with a well-characterised structure both in solution and in the crystal. A good correlation has been found between predicted regions of exposed protein surface and the intensity attenuations induced by the probe on 2D NMR TOCSY cross peaks of tendamistat in the paramagnetic water solution. All the high paramagnetic effects have been interpreted in terms of more efficient competition of TEMPOL with water molecules at some surface positions. The active site of tendamistat coincides with the largest surface patch accessible to the probe. A strong hydration of protein N and C termini can also be suggested by this structural approach, as these locations exhibit reduced paramagnetic perturbations. Provided that the solution structure is known, the use of this paramagnetic probe seems to be well suited to delineate the dynamic behaviour of the protein surface and, more generally, to gain relevant information about the molecular presentation processes.

Fix L, a haemoglobin that acts as an oxygen sensor: signalling mechanism and structural basis of its homology with PAS domains.

Fix L, which contains a haemoglobin domain homologous to the PAS family and a histidine kinase domain, forms, with Fix J, a two-component signalling complex that regulates expression of nitrogenase genes in Rhizobium. Spin transitions of its haem iron trigger stereochemical changes in and around the haem that, together with steric effects, control the activity of the kinase. Homology with the PAS family is based on a common core of about 20 structurally equivalent sites from which polar residues are excluded.

Serpins in the Caenorhabditis elegans genome.

Data mining in genome sequences can identify distant homologues of known protein families, and is most powerful if solved structures are available to reveal the three-dimensional implications of very dissimilar sequences. Here we describe putative serpin sequences identified with very high statistical significance in the Caenorhabditis elegans genome. When mapped onto vertebrate serpins such as alpha1-antitrypsin, they suggest novel structural features. Some appear complete, some show extensive deletions, and others appear to contain only the C-terminal part of the known serpin fold, probably in partnership with N-terminal regions that have conformations unlike those of known serpins. The observation of such striking sequence similarity, in proteins that must have significantly different overall structures, substantially extends the structural characteristics of the serpin family of proteins.

Extraction of geometrically similar substructures: least-squares and Chebyshev fitting and the difference distance matrix.

In analysis, comparison and classification of conformations of proteins, a common computational task involves extractions of similar substructures. Structural comparisons are usually based on either of two measures of similarity: the root-mean-square (r.m.s.) deviation upon optimal superposition, or the maximal element of the difference distance matrix. The analysis presented here clarifies the relationships between different measures of structural similarity, and can provide a basis for developing algorithms and software to extract all maximal common well-fitting substructures from proteins. Given atomic coordinates of two proteins, many methods have been described for extracting some substantial (if not provably maximal) common substructure with low r.m.s. deviation. This is a relatively easy task compared with the problem addressed here, i.e., that of finding all common substructures with r.m.s. deviation less than a prespecified threshold. The combinatorial problems associated with similar subset extraction are more tractable if expressed in terms of the maximal element of the difference distance matrix than in terms of the r.m.s. deviation. However, it has been difficult to correlate these alternative measures of structural similarity. The purpose of this article is to make this connection. We first introduce a third measure of structural similarity: the maximum distance between corresponding pairs of points after superposition to minimize this value. This corresponds to fitting in the Chebyshev norm. Properties of Chebyshev superposition are derived. We describe relationships between the r.m.s. and minimax (Chebyshev) deviations upon optimal superposition, and between the Chebyshev deviation and the maximal element of the difference distance matrix. Combining these produces a relationship between the r.m.s. deviation upon optimal superposition and the maximal element of the difference distance matrix. Based on these results, we can apply algorithms and software for finding subsets of the difference distance matrix for which all elements are less than a specified bound, either to select only subsets for which the r.m.s.deviation is less than or equal to a specified threshold, or to select subsets that include all subsets for which the r.m.s. deviation is less than or equal to a threshold.

Comprehensive, comprehensible, distributed and intelligent databases: current status.

MOTIVATION: It is only a matter of time until a user will see not many but one integrated database of information for molecular biology. Is this true? Is it a good thing? Why will it happen? Where are we now? What developments are fostering and what developments are impeding progress towards this end? SUPPLEMENTARY INFORMATION: A list of WWW resources devoted to database issues in molecular biology is available at http://www.mips.biochem.mpg.de CONTACT: frishman@mips.biochem.mpg.de

An atlas of serpin conformations.

The serpins are a family of proteins that inhibit chymotrypsin-like serine proteinases, with an unusual mechanism involving a large conformational change known as the stressed-->relaxed (S-->R) transition. This article is a guide to the known serpin conformations and their biological significance.

Conformations of the third hypervariable region in the VH domain of immunoglobulins.

Antigen-combining sites of antibodies are constructed from six loops from VL and VH domains. The third hypervariable region of the heavy chain is far more variable than the others in length, sequence and structure, and was not included in the canonical-structure description of the conformational repertoire of the three hypervariable regions of V kappa chains and the first two of VH chains. Here we present an analysis of the conformations of the third hypervariable region of VH domains (the H3 regions) in antibodies of known structure. We define the H3 region as comprising the residues between 92Cys and 104Gly. We divide it into a torso comprising residues proximal to the framework, four residues from the N terminus and six residues from the C terminus, and a head. There are two major classes of H3 structures that have more than ten residues between 92Cys and 104Gly: (1) the conformation of the torso has a beta-bulge at residue 101, and (2) the torso does not contain a bulge, but continues the regular hydrogen-bonding pattern of the beta-sheet hairpin. The choice of bulged versus non-bulged torso conformation is dictated primarily by the sequence, through the formation of a salt bridge between the side-chains of an Arg or Lys at position 94 and an Asp at position 101. Thus the torso region appears to have a limited repertoire of conformations, as in the canonical structure model of other antigen-binding loops. The heads or apices of the loops have a very wide variety of conformations. In shorter H3 regions, and in those containing the non-bulged torso conformation, the heads follow the rules relating sequence to structure in short hairpins. We surveyed the heads of longer H3 regions, finding that those with bulged torsos present many very different conformations of the head. We recognize that H3, unlike the other five antigen-binding loops, has a conformation that depends strongly on the environment, and we have analysed the interactions of H3 with residues elsewhere in the VH domain, in the VL domain, and with ligands, and their effects on the conformation of H3. We tested these results by attempts to predict the conformations of H3 regions in antibody structures solved after the results were derived. The general conclusion of this work is that the conformation of H3 shows some regularities, from which rules relating sequence to conformation can be stated, but to a less complete degree than for the other five antigen-binding loops. Accurate prediction of the torso conformation is possible in most cases; predictions of the conformation of the head is possible in some cases. However, our understanding of the sequence-structure relationships has reduced the uncertainty to no more than a few residues at the apex of the H3 region.

Preparative induction and characterization of L-antithrombin: a structural homologue of latent plasminogen activator inhibitor-1.

The inhibitory mechanism of the serpin family of serine protease inhibitors is characterized by a remarkable degree of conformational flexibility. Various conformational states have been elucidated by X-ray crystallography and indicate that the inhibitory loop, the central A-beta-sheet, and the outside edge of the C-beta-sheet are particularly mobile. However, no crystal structure of a serpin-enzyme complex is yet available, and the likely nature of the protease-complexed serpin remains for biochemical and biophysical researchers to examine. Here, we show that the biochemical induction of the latent state of antithrombin is slow relative to polymer formation, and infer that this may reflect structural features that are important for the regulation of the initial docking and subsequent locking of serpins with cognate proteases. L-Antithrombin was induced by incubation of native antithrombin at 60 degrees C for 10 h in the presence of citrate to prevent polymerization. L-Antithrombin was more stable to denaturation by both heat and urea than native antithrombin. Whereas native antithrombin formed binary complexes with synthetic peptide homologues of the inhibitory loop, biochemically induced L-antithrombin did not, indicating that the inhibitory loop of L-antithrombin is probably fully inserted into the A-beta-sheet as in the crystal structure. This was confirmed by limited proteolysis studies which demonstrated that the inhibitory loop of L-antithrombin could not be cleaved by five proteases which do cleave the loop of native antithrombin. The limited proteolysis studies also indicated that the "gate" region (residues 236-248) of the biochemically induced L-antithrombin was in a conformation substantially different from that of the native antithrombin. This again is similar to L-antithrombin in the crystal structure in which the gate has "opened" away from the body of the molecule by a rotation of 24 degrees to facilitate the relocation of strand 1C from its ordered position in the C-beta-sheet to a disordered surface loop. At 60 degrees C in the absence of citrate, antithrombin (and other serpins) rapidly polymerizes. In the presence of citrate, the formation of L-antithrombin is slow and increases with time, indicating that the inhibition of polymer formation by citrate allows the time necessary for the much slower formation of the L form. We therefore suggest that L-antithrombin formation is a two-step process: an initial rapid conformational change, probably including partial incorporation of the reactive loop into the A-sheet (as in the active molecule in the crystal structure) and displacement of s1C from the C-beta-sheet which supports polymer formation, and a much slower transition to complete loop insertion within the A-beta-sheet. It is likely that both the first rapid transitional step and the structural features that impose resistance to the second more extensive conformational change reflect the optimization of the unique inhibitory function in the serpins.

Probing protein structure by solvent perturbation of NMR spectra: the surface accessibility of bovine pancreatic trypsin inhibitor.

In the absence of specific interactions, the relative attenuation of protein NMR signals due to added stable free radicals such as TEMPOL should reflect the solvent accessibility of the molecular surface. The quantitative correlation between observed attenuation and surface accessibility was investigated with a model system, i.e., the small protein bovine pancreatic trypsin inhibitor. A detailed discussion is presented on the reliability and limits of the approach, and guidelines are provided for data acquisition, treatment, and interpretation. The NMR-derived accessibilities are compared with those obtained from x-ray diffraction and molecular dynamics data. Although the time-averaged accessibilities from molecular dynamics are ideally suited to fit the NMR data, better agreement was observed between the paramagnetic attenuations of the fingerprint cross-peaks of homonuclear proton spectra and the total NH and H alpha accessibilities calculated from x-ray coordinates, than from time-averaged molecular dynamics simulations. In addition, the solvent perturbation response appears to be a promising approach for detecting the thermal conformational evolution of secondary structure elements in proteins.