EF-hands at atomic resolution: the structure of human psoriasin (S100A7) solved by MAD phasing.

BACKGROUND: The S100 family consists of small acidic proteins, belonging to the EF-hand class of calcium-binding proteins. They are primarily regulatory proteins, involved in cell growth, cell structure regulation and signal transduction. Psoriasin (S100A7) is an 11.7 kDa protein that is highly upregulated in the epidermis of patients suffering from the chronic skin disease psoriasis. Although its exact function is not known, psoriasin is believed to participate in the biochemical response which follows transient changes in the cellular Ca2+ concentration. RESULTS: The three-dimensional structure of holmium-substituted psoriasin has been determined by multiple anomalous wavelength dispersion (MAD) phasing and refined to atomic resolution (1.05 A). The structure represents the most accurately determined structure of a calcium-binding protein. Although the overall structure of psoriasin is similar to those of other S100 proteins, several important differences exist, mainly in the N-terminal EF-hand motif that contains a distorted loop and lacks a crucial calcium-binding residue. It is these minor differences that may account for the different specificities among members of this family. CONCLUSIONS: The structure of human psoriasin reveals that this protein, in contrast to other S100 proteins with known structure, is not likely to strongly bind more than one calcium ion per monomer. The present study contradicts the idea that calcium binding induces large changes in conformation, as suggested by previously determined structures of apo forms of S100 proteins. The substitution of Ca2+ ions in EF-hands by lanthanide ions may provide a general vehicle for structure determination of S100 proteins by means of MAD phasing.

Dissection of the DNA-binding domain of Xenopus laevis TFIIIA. Quantitative DNase I footprinting analysis of specific complexes between a 5 S RNA gene fragment and N-terminal fragments of TFIIIA containing three, four or five zinc-finger domains.

Recombinant zinc finger proteins corresponding to N-terminal fragments of Xenopus laevis transcription factor IIIA (TFIIIA) comprising three, four and five fingers produced in Escherichia coli as cleavable hybrid proteins were shown to form specific stoichiometric complexes with DNA fragments containing the internal control region (ICR) of a 5 S RNA gene. The ordered set of DNase I footprints of each of the three proteins on the ICR comprise a nested set of footprints extending upstream from its 3' end (position +96 relative to start of the mature transcript) 20 bp, 20 bp or 34 bp into the ICR, respectively. Quantitative analysis of the footprinting data provided firm evidence that the DNase I footprint, and hence the structure, of the authentic TFIIIA:ICR complex in this region is fully and precisely accounted for by the N-terminal three fingers binding within the +77 to +96 region plus the pair of fingers 4 and 5, both required to extend the footprint upwards from the +77 to the +63 position. A structural interpretation of this set of new footprinting data in view of previous results and data is presented and discussed in terms of a refined model in which the protein-DNA interaction between the ICR and the three N-terminal fingers corresponds closely to that observed in the homologous three-finger zif268:DNA complex, whereas the basic mode of protein-DNA interaction, in which the pair of fingers 4 and 5 is engaged in forming the TFIIIA:ICR complex is of an entirely different, albeit not yet understood nature. To allow assessment of our model in terms of potential specificity-determining H-bonding patterns, a molecular model of the complex between the three-finger TFIIIA fragment and the ICR was constructed, using the zif268:DNA co-ordinates. Eight out of the nine amino acid residues, which according to our model are suitably located for forming hydrogen bonds with the bases, are potential H-bond acceptors or donors.

Tetranectin, a trimeric plasminogen-binding C-type lectin.

Tetranectin, a plasminogen-binding protein belonging to the family of C-type lectins, was expressed in E. coli and converted to its native form by in vitro refolding and proteolytic processing. Recombinant tetranectin-as well as natural tetranectin from human plasma-was shown by chemical cross-linking analysis and SDS-PAGE to be a homo-trimer in solution as are other known members of the collectin family of C-type lectins. Biochemical evidence is presented showing that an N-terminal domain encoded within exons 1 and 2 of the tetranectin gene is necessary and sufficient to govern subunit trimerization.

Psoriasin binds calcium and is upregulated by calcium to levels that resemble those observed in normal skin.

Recently, we described a small molecular weight protein termed psoriasin that showed sequence similarity with the S100 calcium-binding proteins and that is highly upregulated in psoriatic epidermis as well as in primary human keratinocytes undergoing abnormal differentiation. Here we present evidence showing that natural and recombinant psoriasin binds calcium, as judged by the calcium overlay assay, and that it contains all the sequence features characteristic of the S100 family. Furthermore, [35S]-methionine labeling experiments showed that psoriasin synthesis is upregulated by 2 mM Ca++ (ratio Ca++/control at 88 h = 2.56) to levels that resemble those observed in unfractionated keratinocyte populations obtained from normal skin.

Sequence-specific binding of the N-terminal three-finger fragment of Xenopus transcription factor IIIA to the internal control region of a 5S RNA gene.

An N-terminal fragment of Xenopus TFIIIA, containing domains 1-3, (TF 3), was expressed in E. coli. High yields of recombinant zinc finger protein was isolated, and its DNA binding activity for the internal control region (ICR) of the Xenopus 5S RNA gene, was demonstrated by band-shift experiments and DNase I footprinting analysis. TF 3 protects 20 bp of ICR against DNase I digestion. The limits of protection are from +77 to +96 on both coding and noncoding strand. This protection pattern is identical to the protection pattern obtained with TFIIIA in the overlapping region, showing that the 3-finger fragment accounts fully for the protein-DNA interactions in TFIIIA-5S RNA gene over this region.

Collagen-binding recombinant fibronectin fragments containing type II domains.

Each of the two type II domains and four larger fragments, containing one or two type II domains of fibronectin, have been expressed in Escherichia coli. A special vector, containing a fragment encoding the cleavage site for Factor Xa, Ile-Glu-Gly-Arg, inserted immediately before the protein fragment of interest, was used. After treatment of the purified fusion proteins with reduced/oxidized glutathione, the correctly folded fibronectin fragments were released by proteolytic digestion with Factor Xa. The largest fragment, consisting of two type II and two type I domains, was the only fragment able to bind to immobilized gelatin.

Expression and refolding of a high-affinity receptor binding domain from rat alpha 1-macroglobulin.

A recombinant version of the receptor binding domain of rat alpha 1-macroglobulin (RBDv) consisting of residues 1319-1474 has been expressed in E. coli. Competition experiments with 125I-labelled methylamine treated human alpha 2-macroglobulin reveal that the alpha 1-macroglobulin-RBDv exhibit the same high affinity for the alpha 2-macroglobulin receptor as the entire 40 kDa light chain from rat alpha 1-macroglobulin. It is therefore concluded, that all determinants for receptor interaction reside in the C-terminal approx. 150 residues of the alpha-macroglobulin subunit.

The carboxy-terminal domain of the receptor-associated protein binds to the Vps10p domain of sortilin.

Binding of the receptor-associated protein (RAP) to the newly identified putative sorting receptor, sortilin, was analyzed by surface plasmon resonance analysis of recombinant RAP and sortilin domains and compared with binding to megalin and low density lipoprotein receptor-related protein (LRP). The data show that the RAP-binding site in sortilin is localized in the cysteine-rich lumenal part homologous to yeast vacuolar protein-sorting 10 protein (Vps10p), and the sortilin-binding site in RAP is localized in the carboxy-terminal domain III of the three homologous domains in RAP. Whereas sortilin bound only RAP domain III, megalin and LRP bound all RAP domains with the functional affinity order: domain III >domain I > domain II.

Crystal structure of tetranectin, a trimeric plasminogen-binding protein with an alpha-helical coiled coil.

Tetranectin is a plasminogen kringle 4-binding protein. The crystal structure has been determined at 2.8 A resolution using molecular replacement. Human tetranectin is a homotrimer forming a triple alpha-helical coiled coil. Each monomer consists of a carbohydrate recognition domain (CRD) connected to a long alpha-helix. Tetranectin has been classified in a distinct group of the C-type lectin superfamily but has structural similarity to the proteins in the group of collectins. Tetranectin has three intramolecular disulfide bridges. Two of these are conserved in the C-type lectin superfamily, whereas the third is present only in long-form CRDs. Tetranectin represents the first structure of a long-form CRD with intact calcium-binding sites. In tetranectin, the third disulfide bridge tethers the CRD to the long helix in the coiled coil. The trimerization of tetranectin as well as the fixation of the CRDs relative to the helices in the coiled coil indicate a demand for high specificity in the recognition and binding of ligands.

Receptor-binding domain of human alpha 2-macroglobulin. Expression, folding and biochemical characterization of a high-affinity recombinant derivative.

A recombinant version of the receptor binding domain (RBDv) of human alpha 2-macroglobulin (alpha 2M) has been expressed in E. coli and refolded using a novel iterative procedure. RBDv (Val1299-Ala1451) is extended by 15 residues at the N-terminal side of the Lys1313-Glu papain cleavage site in human alpha 2M. RBDv contains the intra-chain bridge Cys1329-Cys1444 and is soluble and monomeric. Competition experiments with 125I-labelled methylamine-treated alpha 2M reveal that RBDv binds to the placental receptor for transformed alpha 2M with a Kd of 8 nM, i.e. the binding affinity of RBDv is of the same order of magnitude as the intrinsic affinity for binding of one domain in transformed alpha 2M to one receptor molecule.

Nested sets of protein fragments and their use in epitope mapping: characterization of the epitope for the S4D5 monoclonal antibody binding to receptor associated protein.

The present report describes a new general procedure by which linear and some structure-dependent epitopes may be mapped in a protein antigen using a nested set of protein fragments prepared from partial proteolysis products of a recombinant protein. Briefly, the antigen, fused to an affinity tag, is partially fragmented and affinity sorted under denaturing conditions to produce a nested set of polypeptides, consisting of N- (or C-)terminal fragments. Immunoblots of SDS-PAGE fractionated sets of fragments are therefore directly readable in terms of molecular mass--i.e., approximate sequence positions--that identify sequence segments harbouring an epitope and any additional structural elements, required to maintain epitope conformation. Blots of N- and C-terminal nested sets of polypeptide fragments representing the human receptor associated protein (RAP) were prepared and probed with mAb S4D5 (Moestrup and Gliemann, 1991). Fragments 1-177 and 94-323 were the shortest fragments detected by the antibody, suggesting the presence of an epitope within the 94-177 segment. Independent mapping based on recombinant fragments of the RAP homologue, rat Heymann nephritis antigen, confirmed that the epitope resides in the Pro115-Asp177 segment. The model study demonstrates the utility of nested sets of protein fragments as fast and inexpensive tools for epitope mapping.

Amino acid sequence of hen ovomacroglobulin (ovostatin) deduced from cloned cDNA.

The complete amino acid sequence of the hen ovomacroglobulin (ovostatin) subunit has been determined from cDNA and partial peptide sequence analysis. Ovostatin is a tetrameric member of the alpha-macroglobulin (alpha M) family of proteins. The 4715 nt ovostatin cDNA encodes a 36- or a 16-residue signal peptide and a 1437-residue mature protein (162.2 kDa). At the protein level the overall score of sequence identity between ovostatin and mammalian alpha Ms is 39-44%, indicating an early divergence from the line leading to the mammalian alpha Ms. Ovostatin contains 56 mol glucosamine per mol subunit, and 12 of its Asn-residues are likely to be N-glycosylated. Including carbohydrate, the size of the ovostatin subunit is approx. 185 kDa. The ovostatin subunit is predicted to contain 12 intrachain disulfide bridges, and two subunits are predicted to be disulfide bound by two interchain bridges. One Cys residue may be unpaired or participate in dimer formation as a third interchain disulfide bridge. Ovostatin contains a unique 40-residue bait region. In contrast to other alpha Ms, ovostatin contains no internal beta-Cys-gamma-Glu thiol ester, as a result of a Cys-to-Asn replacement (TGC or TGT to AAT), but the Gln-moiety of the thiol ester is preserved. By comparing the sequences of the receptor binding domain in alpha Ms with the corresponding region of ovostatin possible determinants for receptor recognition of mammalian alpha Ms are proposed.

Generation of beta-globin by sequence-specific proteolysis of a hybrid protein produced in Escherichia coli.

High-level expression of many eukaryotic genes has proved difficult to achieve even when a strong promoter and the ribosome binding sequence from highly expressed Escherichia coli genes have been placed in front of the coding sequences. To overcome this problem, many eukaryotic proteins have been efficiently produced as hybrids after fusion of their genes with a coding sequence of E. coli genes. However, such hybrid proteins are not suitable for functional studies or clinical use unless the authentic protein sequence can be released by specific cleavage. Here, we have inserted the sequence Ile-Glu-Gly-Arg between the 31 amino-terminal residues of lambda cII protein and Val 1 of human beta-globin, and produced this hybrid in high yield in E. coli. We then cleaved the hybrid specifically at the single arginine, using blood coagulation factor Xa and thus liberated the authentic beta-globin chain. As factor Xa is specific for the tetrapeptide Ile-Glu-Gly-Arg, which is rare in protein sequences, our expression/cleavage system is applicable to the efficient production of many eukaryotic proteins.

Complete primary structure of the collagen-binding domain of bovine fibronectin.

The complete amino acid sequence of the collagen-binding domain of bovine plasma fibronectin has been determined. The fragment, generated by digestion of fibronectin with plasmin and chymotrypsin, contains 340 residues (260-599 of fibronectin) with threonine and tryptophan as the amino-terminal and carboxyl-terminal amino acids, respectively. 24 half-cystines and no cysteines are present in the sequence. Three glucosamine-based oligosaccharide groups are attached to Asn-399, Asn-497 and to Asn-511, respectively. Two of the three types (I and II) [Petersen et al. (1983) Proc. Natl Acad. Sci. USA 80, 137-141] of internal homology occur in the fragment, namely four of the at least twelve stretches of type I sequence homology, 'fingers', and two stretches of type II homology. The type I homology is present in two other plasmic fragments from fibronectin, while the type II homology has been found in the collagen-binding domain only.

Cloning and sequencing of the immunoglobulin A1 protease gene (iga) of Haemophilus influenzae serotype b.

Secretion of immunoglobulin A1 (IgA1) proteases is a characteristic of Haemophilus influenzae and several other bacterial pathogens causing infectious diseases, including meningitis. Indirect evidence suggests that the proteases are important virulence factors. In this study, we cloned the iga gene encoding immunoglobulin A1 (IgA1) protease from H. influenzae serotype b into Escherichia coli, in which the recombinant H. influenzae iga gene was expressed and the resulting protease was secreted. Sequencing a part of a 7.5-kilobase DNA fragment containing the iga gene revealed a large open reading frame with a strongly biased codon usage and having the potential of encoding a protein of 1,541 amino acids and a molecular mass of 169 kilodaltons. Putative promoter and terminator elements flanking the open reading frame were identified. Comparison of the deduced amino acid sequence of this H. influenzae IgA1 protease with that of a similar protease from Neisseria gonorrhoeae revealed several domains with a high degree of homology. Analogous to mechanisms known from the N. gonorrhoeae IgA protease secretion, we propose a scheme of posttranslational modifications of the H. influenzae IgA1 protease precursor, leading to a secreted protease with a molecular mass of 108 kilodaltons, which is close to the 100 kilodaltons reported for the mature IgA1 protease.

The solution structure of the N-terminal domain of alpha2-macroglobulin receptor-associated protein.

The three-dimensional structure of the N-terminal domain (residues 18-112) of alpha2-macroglobulin receptor-associated protein (RAP) has been determined by NMR spectroscopy. The structure consists of three helices composed of residues 23-34, 39-65, and 73-88. The three helices are arranged in an up-down-up antiparallel topology. The C-terminal 20 residues were shown not to be in a well defined conformation. A structural model for the binding of RAP to the family of low-density lipoprotein receptors is proposed. It defines a role in binding for both the unordered C terminus and the structural scaffold of the core structure. Pathogenic epitopes for the rat disease Heymann nephritis, an experimental model of human membranous glomerulonephritis, have been identified in RAP and in the large endocytic receptor gp330/megalin. Here we provide the three-dimensional structure of the pathogenic epitope in RAP. The amino acid residues known to form the epitope are in a helix-loop-helix conformation, and from the structure it is possible to rationalize the published results obtained from studies of fragments of the N-terminal domain.

Crystallization and preliminary X-ray diffraction studies of psoriasin.

Crystals of psoriasin, a protein related to the skin disease psoriasis, have been grown in two different crystal forms. Form I represents the protein in the Ca(2+)-bound form, and form II represents the protein in the Zn(2+)- and Ca(2+)-bound form. The crystals of form I are orthorhombic belonging to the space group P2(1)2(1)2(1) with cell parameters a = 52.15, b = 56.67 and c = 76.38 A and diffract to 2.4 A. The crystals of form II are tetragonal and belong to the space group P4(1(3))2(1)2 with cell parameters a = b = 51.86, c = 115.93 A and diffract to 2.0 A.