Anthranilate sulfonamide hydroxamate TACE inhibitors. Part 2: SAR of the acetylenic P1' group.

The SAR of a series of potent sulfonamide hydroxamate TACE inhibitors bearing novel acetylenic P1' groups was explored. In particular, compound 4t bearing a butynyloxy P1' moiety has excellent in vitro potency against isolated TACE enzyme and in cells, good selectivity over MMP-1 and oral activity in an in vivo model of TNF-alpha production.

The discovery of anthranilic acid-based MMP inhibitors. Part 3: incorporation of basic amines.

Anthranilic acid derivatives bearing basic amines were prepared and evaluated in vitro and in vivo as inhibitors of MMP-1, MMP-9, MMP-13, and TACE. Piperazine 4u has been identified as a potent, selective, orally active inhibitor of MMP-9 and MMP-13.

The discovery of anthranilic acid-based MMP inhibitors. Part 2: SAR of the 5-position and P1(1) groups.

A novel series of anthranilic acid-based inhibitors of MMP-1, MMP-9, MMP-13, and TACE was prepared and evaluated. Selective inhibitors of MMP-9, MMP-13, and TACE were identified, including the potent, orally active MMP-13 inhibitor 4p.

An essential role for ectodomain shedding in mammalian development.

The ectodomains of numerous proteins are released from cells by proteolysis to yield soluble intercellular regulators. The responsible protease, tumor necrosis factor-alpha converting enzyme (TACE), has been identified only in the case when tumor necrosis factor-alpha (TNFalpha) is released. Analyses of cells lacking this metalloproteinase-disintegrin revealed an expanded role for TACE in the processing of other cell surface proteins, including a TNF receptor, the L-selectin adhesion molecule, and transforming growth factor-alpha (TGFalpha). The phenotype of mice lacking TACE suggests an essential role for soluble TGFalpha in normal development and emphasizes the importance of protein ectodomain shedding in vivo.

Structure-function analysis of FLT3 ligand-FLT3 receptor interactions using a rapid functional screen.

FLT3 ligand (FLT3L) stimulates primitive hematopoietic cells by binding to and activating the FLT3 receptor (FLT3R). We carried out a structure-activity study of human FLT3L in order to define the residues involved in receptor binding. We developed a rapid method to screen randomly mutagenized FLT3L using a FLT3R-Fc fusion protein to probe the relative binding activities of mutated ligand. Approximately 60,000 potential mutants were screened, and the DNA from 59 clones was sequenced. Thirty-one single amino acid substitutions at 24 positions of FLT3L either enhanced or reduced activity in receptor binding and cell proliferation assays. Eleven representative proteins were purified and analyzed for receptor affinity, specific activity, and physical properties. Receptor affinity and bioactivity were highly correlated. FLT3L affinity for receptor improved when four individual mutations that enhance FLT3L receptor affinity were combined in a single molecule. A model of FLT3L three-dimensional structure was generated based on sequence alignment and x-ray structure of macrophage colony-stimulating factor. Most residues implicated in receptor binding are widely dispersed in the primary structure of FLT3L, yet they localize to a surface patch in the tertiary model. A mutation that maps to and is predicted to disrupt the proposed dimerization interface between FLT3L monomers exhibits a Stokes radius that is concentration-dependent, suggesting that this mutation disrupts the FLT3L dimer.

Crystal structure of the catalytic domain of human tumor necrosis factor-alpha-converting enzyme.

Tumor necrosis factor-alpha (TNFalpha) is a cytokine that induces protective inflammatory reactions and kills tumor cells but also causes severe damage when produced in excess, as in rheumatoid arthritis and septic shock. Soluble TNFalpha is released from its membrane-bound precursor by a membrane-anchored proteinase, recently identified as a multidomain metalloproteinase called TNFalpha-converting enzyme or TACE. We have cocrystallized the catalytic domain of TACE with a hydroxamic acid inhibitor and have solved its 2.0 A crystal structure. This structure reveals a polypeptide fold and a catalytic zinc environment resembling that of the snake venom metalloproteinases, identifying TACE as a member of the adamalysin/ADAM family. However, a number of large insertion loops generate unique surface features. The pro-TNFalpha cleavage site fits to the active site of TACE but seems also to be determined by its position relative to the base of the compact trimeric TNFalpha cone. The active-site cleft of TACE shares properties with the matrix metalloproteinases but exhibits unique features such as a deep S3' pocket merging with the S1' specificity pocket below the surface. The structure thus opens a different approach toward the design of specific synthetic TACE inhibitors, which could act as effective therapeutic agents in vivo to modulate TNFalpha-induced pathophysiological effects, and might also help to control related shedding processes.

A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells.

Mammalian cells proteolytically release (shed) the extracellular domains of many cell-surface proteins. Modification of the cell surface in this way can alter the cell's responsiveness to its environment and release potent soluble regulatory factors. The release of soluble tumour-necrosis factor-alpha (TNF-alpha) from its membrane-bound precursor is one of the most intensively studied shedding events because this inflammatory cytokine is so physiologically important. The inhibition of TNF-alpha release (and many other shedding phenomena) by hydroxamic acid-based inhibitors indicates that one or more metalloproteinases is involved. We have now purified and cloned a metalloproteinase that specifically cleaves precursor TNF-alpha. Inactivation of the gene in mouse cells caused a marked decrease in soluble TNF-alpha production. This enzyme (called the TNF-alpha-converting enzyme, or TACE) is a new member of the family of mammalian adamalysins (or ADAMs), for which no physiological catalytic function has previously been identified. Our results should facilitate the development of therapeutically useful inhibitors of TNF-alpha release, and they indicate that an important function of adamalysins may be to shed cell-surface proteins.

Resonance assignments for Oncostatin M, a 24-kDa alpha-helical protein.

Oncostatin M (OM) is a cytokine that shares a structural and functional relationship with interleukin-6, leukemia inhibitory factor, and granulocyte-colony stimulating factor, which regulate the proliferation and differentiation of a variety of cell types. A mutant version of human OM in which two N-linked glycosylation sites and an unpaired cysteine have been mutated to alanine (N76A/C81A/N193A) has been expressed and shown to be active. The triple mutant has been doubly isotope-labeled with 13C and 15N in order to utilize heteronuclear multidimensional NMR techniques for structure determination. Approximately 90% of the backbone resonances were assigned from a combination of triple-resonance data (HNCA, HNCO, CBCACONH, HBHACONH, HNHA and HCACO), intraresidue and sequential NOEs (3D 15N-NOESY-HMQC and 13C-HSQC-NOESY) and side-chain information obtained from the CCONH and HCCONH experiments. Preliminary analysis of the NOE pattern in the 15N-NOESY-HMQC spectrum and the 13C alpha secondary chemical shifts predicts a secondary structure for OM consisting of four alpha-helices with three intervening helical regions, consistent with the four-helix-bundle motif found for this cytokine family. As a 203-residue protein with a molecular weight of 24 kDa, Oncostatin M is the largest alpha-helical protein yet assigned.

The human homologue of rat NG2, a chondroitin sulfate proteoglycan, is not expressed on the cell surface of normal hematopoietic cells but is expressed by acute myeloid leukemia blasts from poor-prognosis patients with abnormalities of chromosome band 11q23.

In our efforts to produce monoclonal antibodies that recognize cell-surface antigens expressed by hematopoietic precursor and stromal cells, we generated a monoclonal antibody, 7.1, which recognizes a 220- to 240-kD cell-surface protein whose N-terminal amino acid sequence is identical to the rat NG2 chondroitin sulfate proteoglycan molecule. This chondroitin sulfate proteoglycan, previously reported to be expressed by human melanoma cells, was not found to be expressed by normal hematopoietic cells, nor was it expressed on the cell surface of cell lines of hematopoietic origin including cell lines with 11q23 abnormalities. It was found on the cell surface of acute myeloid leukemia (AML) blasts and cell lines derived from nonhematopoietic tissues. Samples of leukemic marrow from 166 children with AML enrolled on Childrens Cancer Group protocol 213 were evaluated for cell-surface expression of this proteoglycan molecule. In 18 of 166 (11%) patient samples, greater than 25% of leukemic blasts expressed the NG2 molecule. These 18 patients had a poorer outcome with respect to survival (P = .002) and event-free survival (P = .035) with an actuarial survival at 4 years of 16.7%. Blast cell expression of the NG2 molecule was strongly associated with French-American-British M5 morphology (P < .0001) and abnormalities in chromosome band 11q23, site of the MLL gene. These results show that the NG2 molecule is expressed by malignant hematopoietic cells that have abnormalities in chromosome band 11q23, suggesting that antibody 7.1 may be useful in the rapid identification of this group of poor-prognosis patients.

Modeling protein loops using a phi i + 1, psi i dimer database.

We present an automated method for modeling backbones of protein loops. The method samples a database of phi i + 1 and psi i angles constructed from a nonredundant version of the Protein Data Bank (PDB). The dihedral angles phi i + 1 and psi i completely define the backbone conformation of a dimer when standard bond lengths, bond angles, and a trans planar peptide configuration are used. For the 400 possible dimers resulting from 20 natural amino acids, a list of allowed phi i + 1, psi i pairs for each dimer is created by pooling all such pairs from the loop segments of each protein in the nonredundant version of the PDB. Starting from the N-terminus of the loop sequence, conformations are generated by assigning randomly selected pairs of phi i + 1, psi i for each dimer from the respective pool using standard bond lengths, bond angles, and a trans peptide configuration. We use this database to simulate protein loops of lengths varying from 5 to 11 amino acids in five proteins of known three-dimensional structures. Typically, 10,000-50,000 models are simulated for each protein loop and are evaluated for stereochemical consistency. Depending on the length and sequence of a given loop, 50-80% of the models generated have no stereochemical strain in the backbone atoms. We demonstrate that, when simulated loops are extended to include flanking residues from homologous segments, only very few loops from an ensemble of sterically allowed conformations orient the flanking segments consistent with the protein topology. The presence of near-native backbone conformations for loops from five different proteins suggests the completeness of the dimeric database for use in modeling loops of homologous proteins. Here, we take advantage of this observation to design a method that filters near-native loop conformations from an ensemble of sterically allowed conformations. We demonstrate that our method eliminates the need for a loop-closure algorithm and hence allows for the use of topological constraints of the homologous proteins or disulfide constraints to filter near-native loop conformations.

Direct evidence of a heterotrimeric complex of human interleukin-4 with its receptors.

The mode of binding of interleukin-4 (IL-4) to its two known receptors, specific receptor IL-4R and a shared receptor gamma c, was investigated using gel filtration and gel electrophoresis. A ternary complex between IL-4 and the soluble domains of the two receptors was shown to exist in solution. The association constant between gamma c and the stable complex of IL-4/sIL-4R is in the millimolar range, making the ternary complex a feasible target for crystallization studies.

Homology modeling of divergent proteins.

A method is presented for homology modeling of proteins bearing weak sequence identity to proteins of known tertiary structure. To accommodate non-identical amino acids in the core region, the backbone of the structurally conserved core of the model protein is allowed to deviate from that of the template protein. We have expanded FOLDER, a distance geometry-based homology modeling method, to allow for such displacements in the structurally conserved core. Models are built by rigidly constraining the interatomic distances within a structurally conserved segment and by allowing the interatomic distances between these segments to vary by a "divergence factor". We test this method by simulating models of the beta-barrel domain D1 of CD4 and a four-helix bundle protein cytochrome b562 using the crystal structures of Bence-Jones protein and cytochrome c' as templates, respectively. In both cases, previously published structure-based sequence alignments were used for simulating models. The root-mean-square (r.m.s.) deviation of the backbone atoms in the common core between the templates and models was found to be a function of the imposed divergence factor. Our results demonstrate that this r.m.s. deviation results from the relative displacements of structurally conserved segments to accommodate the amino acid replacements in the core of the model protein. To test the integrity of the simulated structures we compared them with their respective crystal structures. The r.m.s. deviation of the backbone atoms in the core regions of the simulated models and their respective crystal structures is approximately 1.4 A. The r.m.s. deviation for all the backbone atoms in the models, including those in the structurally variable regions, which are modeled de novo, is 2.4 A for CD4 and 3.2 A for cytochrome b562 when compared with their respective X-ray structures.

Stoichiometry of the complex of human interleukin-4 with its receptor.

A large number of cytokines have been shown to possess a four-helix bundle structure with a unique up-up-down-down connectivity. The receptors for this family of cytokines have been shown to be homologous as well, each possessing two tandem repeats of a fibronectin type III-like domain. The crystal structure of human growth hormone bound to the soluble portion of its receptor has served as the only experimentally-determined example of the interaction between the four-helix bundle cytokines and their receptors: two identical receptor subunits bind to different epitopes on the same growth hormone ligand. We have conducted a series of experiments to determine if this structural paradigm is true for interleukin-4 and interleukin-4 receptor. Native polyacrylamide gel electrophoresis and gel filtration chromatography reveal that interleukin-4 forms a tight 1:1 complex with the system.

The high-resolution, three-dimensional solution structure of human interleukin-4 determined by multidimensional heteronuclear magnetic resonance spectroscopy.

The high-resolution three-dimensional solution structure of recombinant human interleukin-4 (IL-4), a protein of approximately 15 kDa which plays a key role in the regulation of B and T lymphocytes, has been determined using three- and four-dimensional heteronuclear NMR spectroscopy. The structure is based on a total of 2973 experimental NMR restraints, comprising 2515 approximate interproton distance restraints, 102 distance restraints for 51 backbone hydrogen bonds, and 356 torsion angle restraints. A total of 30 structures was calculated by means of hybrid distance geometry-simulated annealing, and the atomic rms distribution about the mean coordinate positions for residues 8-129 is 0.44 +/- 0.03 A for the backbone atoms, 0.83 +/- 0.03 A for all atoms, and 0.51 +/- 0.04 A for all atoms excluding disordered side chains. The N- and C-terminal residues (1-7 and 130-133, respectively) appear to be disordered. The structure of IL-4 is dominated by a left-handed four-helix bundle with an unusual topology comprising two overhand connections. The linker elements between the helices are formed by either long loops, small helical turns, or short strands. The latter include a mini anti-parallel beta-sheet. A best fit superposition of the NMR structure of IL-4 with the 2.25 A resolution crystal structure [Wlodawer, A., Pavlovsky, A., & Gutschina, A. (1992) FEBS Lett. 309, 59-64] yields a backbone atomic rms difference of 1.37 A which can be mainly attributed to tighter packing of the helices in the crystal structure. This is indicated by an approximately 20% reduction in the axial separation of three pairs of helices (alpha A-alpha C, alpha A-alpha D, and alpha C-alpha D) in the crystal structure relative to the NMR structure and may reflect the greater flexibility of the molecule in solution which is reduced in the crystal due to intermolecular contacts. Comparison of the NMR structure of IL-4 with the X-ray structures of two other related proteins, granulocyte-macrophage colony stimulating factor [Diedrichs, K., Boone, T., & Karplus, P. A. (1992) Science 254, 1779-1782] and human growth hormone [de Vos, A. M., Ultsch, M., & Kossiakoff, A. A. (1992) Science 255, 306-312], that bind to the same hematopoietic superfamily of cell surface receptors reveals a remarkably similar topological fold, despite the absence of any significant overall sequence identity, and substantial differences in the relative lengths of the helices, the lengths and the nature of the various connecting elements, and the pattern and number of disulfide bridges.(ABSTRACT TRUNCATED AT 400 WORDS)

Identification of a distinct low-affinity receptor for human interleukin-4 on pre-B cells.

Biotinylated interleukin-4 (IL-4) was used to examine IL-4 receptor (IL-4R) expression on a range of human B-cell lines by flow cytometry. Using high concentrations of biotinylated IL-4, we have identified a novel low-affinity IL-4 receptor expressed at high levels on pre-B lines. Expression of this low-affinity receptor did not correlate with detected mRNA levels for the previously cloned receptor or with reactivity of two anti-human IL-4R monoclonal antibodies (MoAb). Radiolabeled IL-4 cross-linking studies using pre-B lines showed a doublet of 65 to 75 Kd in contrast to the 110- to 130-Kd molecule detected on cells expressing the cloned IL-4R. A soluble IL-4 binding protein (IL-4bp) was purified from the supernatants of three pre-B lines expressing the low-affinity receptor on their surface. IL-4bp could block both IL-4-mediated CD23 induction on tonsil B cells and IL-4-induced inhibition of proliferation of the pre-B line JM1. Partial N-terminal amino acid sequence was obtained from purified IL-4bp that confirmed this protein to be novel. A 12 amino acid peptide based on the IL-4bp sequence was used to produce a polyclonal antiserum that was reactive with purified IL-4bp, and also bound to the surface of pre-B cells but not to murine CTLL cells transfected with the human IL-4R. Blocking MoAb against the previously characterized high-affinity receptor inhibited IL-4-mediated proliferation of hIL-4R+ CTLL cells but had no effect on IL-4-induced inhibition of JM1 cell proliferation, and only partially inhibited IL-4-mediated CD23 and sIgM induction and proliferation of tonsil B cells. The data presented here provide evidence for a novel cell-surface expressed low-affinity IL-4R that also exists as a biologically active soluble IL-4 binding protein.

An automated method for modeling proteins on known templates using distance geometry.

We present an automated method incorporated into a software package, FOLDER, to fold a protein sequence on a given three-dimensional (3D) template. Starting with the sequence alignment of a family of homologous proteins, tertiary structures are modeled using the known 3D structure of one member of the family as a template. Homologous interatomic distances from the template are used as constraints. For nonhomologous regions in the model protein, the lower and the upper bounds for the interatomic distances are imposed by steric constraints and the globular dimensions of the template, respectively. Distance geometry is used to embed an ensemble of structures consistent with these distance bounds. Structures are selected from this ensemble based on minimal distance error criteria, after a penalty function optimization step. These structures are then refined using energy optimization methods. The method is tested by simulating the alpha-chain of horse hemoglobin using the alpha-chain of human hemoglobin as the template and by comparing the generated models with the crystal structure of the alpha-chain of horse hemoglobin. We also test the packing efficiency of this method by reconstructing the atomic positions of the interior side chains beyond C beta atoms of a protein domain from a known 3D structure. In both test cases, models retain the template constraints and any additionally imposed constraints while the packing of the interior residues is optimized with no short contacts or bond deformations. To demonstrate the use of this method in simulating structures of proteins with nonhomologous disulfides, we construct a model of murine interleukin (IL)-4 using the NMR structure of human IL-4 as the template. The resulting geometry of the nonhomologous disulfide in the model structure for murine IL-4 is consistent with standard disulfide geometry.

Purification of interleukin-1 beta converting enzyme, the protease that cleaves the interleukin-1 beta precursor.

We have purified the IL-1 beta converting enzyme from the THP-1 cell line using standard chromatographic techniques and obtained the N-terminal amino acid sequence of this novel protein. After stimulation of THP-1 cells with lipopolysaccharide, hydroxyurea, and silica, the protease was solubilized by multiple freeze/thawing. The protein was purified by ion-exchange chromatography, affinity chromatography on blue agarose, gel filtration, and chromatofocusing. The molecular weight of the protein is approximately 22,000 Da and the pI is between 7.1 and 6.8. The overall yield for this procedure was 16% of the activity found in the initial cell lysates. An antiserum raised against a peptide based on the N-terminus was used to precipitate the protease, confirming our identification of the 22,000-Da protein as the IL-1 beta converting enzyme.

An approach to computer-aided inhibitor design: application to cathepsin L.

We have developed an approach to search for molecules that can be used as lead compounds in designing an inhibitor for a given proteolytic enzyme when the 3D structure of a homologous protein is known. This approach is based on taking the cast of the binding pocket of the protease and comparing its dimensions with that of the dimensions of small molecules. Herein the 3D structure of papain is used to model cathepsin L using the comparative modeling technique. The cast of the binding pocket is computed using the crystal structure of papain because the structures of papain and the model of cathepsin L are found to be similar at the binding site. The dimensions of the cast of the binding site of papain are used to screen for molecules from the Cambridge Structural Database (CSD) of small molecules. Twenty molecules out of the 80,000 small molecules in the CSD are found to have dimensions that are accommodated by the papain binding pocket. Visual comparison of the shapes of the cast and the 20 screened molecules resulted in identifying brevotoxin b, a toxin isolated from the 'red tide' dinoflagellate Ptycho brevis (previously classified as Gymonodium breve), as the structure that best fits the binding pocket of papain. We tested the proteolytic activity of papain and cathepsin L in the presence of brevotoxin b and found inhibition of papain and cathepsin L with Kis of 25 microM and 0.6 microM, respectively. We also compare our method with a more elaborate method in the literature, by presenting our results on the computer search for inhibitors of the HIV-1 protease.