Elucidation of cDNA sequences for metallothioneins from rainbow trout, stone loach and pike liver using the polymerase chain reaction.

Metallothionein cDNAs were generated from the livers of three fish species and amplified by PCR. Two distinct coding sequences (A and B) were elucidated for rainbow trout metallothioneins but single isoforms were encoded by genes isolated from the stone loach and pike. Different sized transcripts were observed both with stone loach and pike but these were accounted for by altered lengths of 3' untranslated regions.

Neutrophil adhesion: a point for therapeutic intervention?

It has become increasingly clear over the last 20 years that the potential exists to modulate inflammatory responses with compounds that interfere with intercellular adhesion. This review highlights the adhesion interactions that occur during neutrophil extravasation and indicates some of the possible ways of disrupting these interactions.

Characterization of recombinant gp120 and gp160 from HIV-1: binding to monoclonal antibodies and soluble CD4.

We compared four preparations of recombinant HIV-1 envelope glycoprotein: mammalian (Chinese hamster ovary cells) gp120 (Celltech); baculovirus gp120 from American Biotechnologies Inc. (ABT) and from MicroGeneSys (MGS); and baculovirus gp160 (Institute of Virology, Oxford, UK). Each envelope glycoprotein binds to a neutralizing monoclonal antibody (MAb) directed against the V3 loop, confirming the integrity of this type-specific neutralization epitope. MGS gp120 binds abnormally well to a MAb which recognizes an epitope preferentially exposed on denatured gp120. Consistent with this finding, MGS gp120 binds to soluble CD4 (sCD4) with an affinity 50-100-fold lower than that of Celltech gp120. The affinity of Celltech gp120 from sCD4 is 2.3 nM, indistinguishable from that of gp120 extracted from HIV-1 virions. Baculovirus gp120 (ABT) and gp160 also have a high affinity for sCD4. A significant proportion of anti-gp120 antibodies in HIV-positive human sera recognize epitopes that are dependent on the mammalian glycosylation pattern, and a human HIV-positive serum inhibits the binding of mammalian gp120 to sCD4 five- to 10-fold more potently than it inhibits baculovirus gp120 binding to sCD4.

E-selectin binds to squamous cell carcinoma and keratinocyte cell lines.

E-selectin is an endothelial adhesion molecule that binds carbohydrate epitopes on leukocytes and has been implicated in a potential pathway of tumor metastasis. Keratinocyte cell lines express similar carbohydrate epitopes, one of which, sialyl Lewis X (SL-X) is a ligand for E-selectin and is also expressed by squamous cell carcinomas (SCC) in situ. The functional role of keratinocyte selectin ligands was investigated using a soluble E-selectin chimaeric protein (pE-sel-Ig) containing pig lectin-like and epidermal growth factor-like domains fused to human IgG. After incubation of keratinocyte cell lines (A431 and SVK14) and normal keratinocytes with pE-sel Ig, binding was quantified by flow cytometry. Frozen sections of SCC were overlaid with pE-sel Ig and binding was visualized immunoenzymatically. Immunolabeling was undertaken using monoclonal antibodies (CSLEX-1 and HECA-452), which label E-selectin ligands including sialyl Lewis X. E-selectin bound strongly to A431 and SVK14 cells; the degree of binding paralleled staining intensity with CSLEX-1 antibody. HECA-452 antibody stained A431 cells strongly but SVK14 cells only weakly. Normal keratinocytes and normal epidermis did not express CSLEX-1 or HECA-452 antigens or bind E-selectin. Serial sections of SCC revealed close correlation between fusion protein binding and antibody staining. Antibody pretreatment of tumor sections with CSLEX-1 blocked fusion protein binding, whereas HECA-452 antibody only slightly reduced fusion protein binding. pE-sel Ig pretreated with YT11.1 antibody failed to bind to A431 or SVK14 cells or to SCC. These studies provide functional evidence that SL-X/E-selectin pathways may be important in SCC metastasis and that A431 and SVK14 cells provide a good model to investigate these mechanisms.

The in vitro activity of ADAM-10 is inhibited by TIMP-1 and TIMP-3.

A recombinant soluble form of the catalytic domain of human ADAM-10 was expressed as an Fc fusion protein from myeloma cells. The ADAM-10 was catalytically active, cleaving myelin basic protein and peptides based on the previously described 'metallosheddase' cleavage sites of tumour necrosis factor alpha, CD40 ligand and amyloid precursor protein. The myelin basic protein degradation assay was used to demonstrate that hydroxamate inhibitors of matrix metalloproteinases (MMPs) were also inhibitors of ADAM-10. The natural MMP inhibitors, TIMP-2 and TIMP-4 were unable to inhibit ADAM-10, but TIMP-1 and TIMP-3 were inhibitory. Using a quenched fluorescent substrate assay and ADAM-10 we obtained approximate apparent inhibition constants of 0.1 nM (TIMP-1) and 0.9 nM (TIMP-3). The TIMP-1 inhibition of ADAM-10 could therefore prove useful in distinguishing its activity from that of TACE, which is only inhibited by TIMP-3, in cell based assays.

TNF-alpha converting enzyme (TACE) is inhibited by TIMP-3.

TNF-alpha converting enzyme (TACE; ADAM-17) is a membrane-bound disintegrin metalloproteinase that processes the membrane-associated cytokine proTNF-alpha to a soluble form. Because of its putative involvement in inflammatory diseases, TACE represents a significant target for the design of specific synthetic inhibitors as therapeutic agents. In order to study its inhibition by tissue inhibitors of metalloproteinases (TIMPs) and synthetic inhibitors of metalloproteinases, the catalytic domain of mouse TACE (rTACE) was overexpressed as a soluble Ig fusion protein from NS0 cells. rTACE was found to be well inhibited by peptide hydroxamate inhibitors as well as by TIMP-3 but not by TIMP-1, -2 and -4. These results suggest that TIMP-3, unlike the other TIMPs, may be important in the modulation of pathological events in which TNF-alpha secretion is involved.

Formation of dimeric Fabs in Escherichia coli: effect of hinge size and isotype, presence of interchain disulphide bond, Fab' expression levels, tail piece sequences and growth conditions.

We have made hinge variants of two human Fab's in order to investigate the factors involved in the formation of dimeric Fab's in the periplasm of E. coli. Hinges containing one or more copies of the IgG1 hinge with various numbers of spacing residues were tested. Fab's with hinges based on the gamma2, gamma3 and gamma4 isotypes were also tested. We find that the IgG1 hinge sequence can form approximately 35% F(ab')2 in vivo in shake flask experiments, but that only (approximately) 5% F(ab')2 can be produced during fermentation. IgM and IgA tail-pieces added to Fab's did not effect their multimerisation. The possible role of growth conditions upon F(ab')2 formation in vivo is discussed.

F(ab')2 molecules made from Escherichia coli produced Fab' with hinge sequences conferring increased serum survival in an animal model.

Fab's with hinges based on the human gamma1 sequence containing 1, 2, or 4 cysteines have been produced by high level Escherichia coli periplasmic secretion, and coupled in vitro by reduction/oxidation to form F(ab')2. We find that the F(ab')2 made with hinges containing 2 or 4 cysteines have a high level (approximately 70%) of multiple disulphide bonds. These F(ab')2 molecules have an increased pharmacokinetic stability as measured by area under the curve compared to those made by direct coupling through a single disulphide bond. One particular molecule containing 4 hinge cysteines has a greater pharmacokinetic stability than a F(ab')2 formed by chemical cross-linking. F(ab')2 made from the Fab' with 4 hinge cysteines is also relatively resistant to chemical reduction in vitro allowing partial reduction to expose reactive hinge thiols. These hinge sequences provide a simple method for producing robust F(ab')2 in vitro, obviating the need to use chemical cross-linkers, and provide a route to hinge specific chemical modification with thiol-reactive conjugates.

The V3 loops of the HIV-1 and HIV-2 surface glycoproteins contain proteolytic cleavage sites: a possible function in viral fusion?

Located close to the crown of the V3 type-specific neutralization loop of the human immunodeficiency virus type 1 (HIV-1) (IIIB) SU glycoprotein gp120, are several potential sites that should be susceptible to proteolytic cleavage by enzymes of trypsinlike or chymotrypsinlike specificity, or by aspartic proteinases. The linkages potentially sensitive to chymotryptic/aspartic proteinase cleavage are retained also within the equivalent domain of HIV-2 (ROD) gp105. We show that thrombin and tryptase cleave HIV-1 gp120 specifically at the tryptic site (GPGR decreases AFVT), and that cathepsin E, an endosomal aspartic proteinase, cleaves at the chymotrypsinlike site (GPGRAF decreases VT). HIV-2 gp105 is also cut by cathepsin E at a site (QIML decreases MSGH) in its V3 loop. Cleavage of HIV-1 gp120 by thrombin is enhanced by sCD4 binding, but is prevented by transient exposure of gp120 to nonionic detergent. Thrombin treatment of HIV-1 gp120 destroys the binding sites for some neutralizing monoclonal antibodies (MAbs) on the V3 loop, but does not affect the affinity of gp120 for sCD4. Conversely, binding of neutralizing MAbs to the HIV-1 V3 loop prior to addition of thrombin or cathepsin E blocks the cleavage reactions, and the binding of some HIV-positive sera to gp120 blocks thrombin cleavage. Analysis of published sequences suggests that all HIV-1, HIV-2, and simian immunovirus (SIV) isolates contain potential proteolytic cleavage sites at similar positions in their V3 loops or equivalent domains. We suggest that cleavage of the V3 loop by a cell surface or endosomal proteinase occurs during the HIV-cell fusion reaction, and that neutralizing antibodies directed against the V3 loop might act by inhibition of this reaction.

In vitro and in vivo characterisation of anti-murine IL-13 antibodies recognising distinct functional epitopes.

Interleukin-13 (IL-13) sequentially binds to IL-13Ralpha1 and IL-4Ralpha forming a high affinity signalling complex. This receptor complex is expressed on multiple cell types in the airway and signals through signal transducer and activator of transcription factor-6 (STAT-6) to stimulate the production of chemokines, cytokines and mucus. Antibodies have been generated, using the UCB Selected Lymphocyte Antibody Method (UCB SLAM), that block either binding of murine IL-13 (mIL-13) to mIL-13Ralpha1 and mIL-13Ralpha2, or block recruitment of mIL-4Ralpha to the mIL-13/mIL-13Ralpha1 complex. Monoclonal antibody (mAb) A was shown to bind to mIL-13 with high affinity (K(D) 11 pM) and prevent binding of mIL-13 to mIL-13Ralpha1. MAb B, that also bound mIL-13 with high affinity (K(D) 8 pM), was shown to prevent recruitment of mIL-4Ralpha to the mIL-13/mIL-13Ralpha1 complex. In vitro, mAbs A and B similarly neutralised mIL-13-stimulated STAT-6 activation and TF-1 cell proliferation. In vivo, mAbs A and B demonstrated equipotent, dose-dependent inhibition of eotaxin generation in mice stimulated by intraperitoneal administration of recombinant mIL-13. In an allergic lung inflammation model in mice, mAbs A and B equipotently inhibited muc5ac mucin mRNA upregulation in lung tissue measured two days after intranasal allergen challenge. These data support the design of therapeutics for the treatment of allergic airway disease that inhibits assembly of the high affinity IL-13 receptor signalling complex, by blocking the binding of IL-13 to IL-13Ralpha1 and IL-13Ralpha2, or the subsequent recruitment of IL-4Ralpha.

Molecular cloning of the pyruvate dehydrogenase complex genes of Escherichia coli.

The three components of the pyruvate dehydrogenase complex of Escherichia coli are encoded by three linked genes, ace E (pyruvate dehydrogenase, E1), aceF (dihydrolipoamide acetyltransferase, E2) and lpd (lipoamide dehydrogenase, E3, situated close to the nadC (quinolinate phosphoribosyltransferase) and aroP (general aromatic amino acid permease) genes with the gene order: nadC-aroP-aceE-aceF-lpd. Several types of transducing phages, lambda nadC and lambda lpd, carrying the nadC and lpd genes were isolated from populations of artificially constructed transducing phages containing R.HindIII or R.EcoRI fragments of bacterial DNA, by selecting for their ability to complement the metabolic lesions of the corresponding mutants. The cloned fragments were extended to include a functional ace operon by in vivo methods involving prophage insertion into the nadC-lpd region and aberrant excision to yield lambda nadC-lpd and lambda lpd-ace phages. These contained overlapping segments of bacterial DNA capable of expressing the aceE, aceF and lpd genes. A physical map of a 20 kilobase pairs (kb) segment of bacterial DNA encoding the entire nadC-lpd region, bounded by R.HindIII and R.EcoRI targets and possessing several internal restriction targets, R.HindIII (3) and R.EcoRI (2), was constructed. Using a combination of nutritional and enzymological studies with dilysogens and genetic analysis with ace mutants the approximate positions of the genes specifying the pyruvate dehydrogenase complex were traced to a 9.5 kb segment of the restriction map. The cloned lpd gene was expressed in the complete absence of a functional ace operon and when the major lambda promoters were repressed. This confirms that the lpd gene can be independently transcribed from its own promoter.

Hybrid plasmids containing the pyruvate dehydrogenase complex genes and gene-DNA relationships in the 2 to 3 minute region of the Escherichia coli chromosome.

A sample of colonies from the Clarke-Carbon ColE1-Escherichia coli DNA plasmid gene bank was screened by conjugation for complementation of the lipoamide dehydrogenase lesion of a deletion strain lacking all components of the pyruvate dehydrogenase complex, delta (aroP aceE aceF lpd). Two ColE1-lpd+ hybrid plasmids were identified: pGS2 (ColE1-ace lpd+; 24 kb) and pGS5 (ColE1-lpd+; 14 kb). Enzymological studies confirmed that pGS2 expressed all the activities of the pyruvate dehydrogenase complex, whereas pGS5 expressed the lipoamide dehydrogenase and acetyltransferase activities (the latter from a ColE1 promoter). These and other plasmids were used to construct a 47-site (15 enzymes) restriction map for a 24.2 kb segment of bacterial DNA in the nadC-lpd region. A further 13 sites (six enzymes) were defined in a 5.4 kb sub-segment containing the lpd gene. lambda phage derivatives containing specific fragments were constructed and used in transduction studies which located the ace and lpd genes in a 7.78 kb sub-segment flanked by AccI and NruI sites.

Stromelysin is an activator of procollagenase. A study with natural and recombinant enzymes.

The latent forms of stromelysin and collagenase from human gingival fibroblasts were purified to homogeneity. These latent proenzymes underwent serial small reductions in Mr upon activation by treatment with either 4-aminophenylmercuric acetate or trypsin. Similar shifts in Mr and activation kinetics were observed upon identical treatments of either recombinant prostromelysin or procollagenase. Prostromelysin showed a lag between activation and Mr fall, suggesting an initial activation by conformational change. Collagenase activity was enhanced up to 12-fold by either natural or recombinant stromelysin in the presence of trypsin or 4-aminophenylmercuric acetate. Stromelysin caused a further apparent decrease in the Mr of procollagenase. Since these important connective-tissue-degrading enzymes are usually co-ordinately produced by cells, a cascade mechanism is proposed in which collagenase is activated by stromelysin.

Mouse cell lines that use heat shock promoters to regulate the expression of tissue plasminogen activator.

The promoters from Drosophila and human 70,000-dalton heat shock protein (hsp70) genes were linked to human tissue plasminogen activator (tPA) cDNA. Mouse C127 cells were transformed with bovine papilloma virus (BPV) vectors carrying the hybrid hsp70/tPA genes. Stable BPV-transformed cell lines were selected and analyzed for tPA expression before and after heat shock. In most cell lines, there was a low level of tPA production even in the absence of heat shock or other obvious stress. After heat shock (42 degrees C, 2 hr), there was up to a 40-fold increase in tPA production. Production of tPA protein occurred within the first 5 h after the heat shock and was due to a burst of hsp70/tPA transcription during the heat shock. The hsp70/tPA transcripts appeared to have a short half-life. Thus, stable mouse cell lines carrying hsp70/tPA hybrid genes can be induced by a short heat shock to transcribe high levels of hsp70/tPA mRNAs and, subsequently, to produce elevated levels of tPA protein.

Antibodies that activate beta 2 integrins can generate different ligand binding states.

A human erythroleukemic cell line (K562) that does not normally express beta 2 integrins has been transfected with the genes encoding these integrins. The resulting cell lines show minimal background adhesion but can be stimulated to bind to appropriate substrates when activated with either of two different antibodies to CD18. The two antibodies appear to generate different ligand binding states in LFA-1 such that different members of the ICAM family are recognized. Antibody-activated complement receptor type 3 and p150,95-transfected cells bind protein-coated surfaces, although they require slightly different activation conditions for optimal binding.

Expression of a soluble functional form of the integrin alpha4beta1 in mammalian cells.

The integrin alpha4beta1(VLA4) has been expressed as a soluble, active, heterodimeric immunoglobulin fusion protein. cDNAs encoding the extracellular domains of the human alpha4 and beta1 subunits were fused to the genomic DNA encoding the human gamma1 immunoglobulin Fc domain and functional integrin fusion protein was expressed as a secreted, soluble molecule from a range of mammalian cell lines. Specific mutations were introduced into the Fc region of the molecules to promote alpha4beta1 heterodimer formation. The soluble alpha4beta1-Fc fusion protein exhibited divalent cation dependent binding to VCAM-1, which was blocked by the appropriate function blocking antibodies. The apparent Kd for VCAM-1 binding were similar for both the soluble and native forms of alpha4beta1. In addition, the integrin-Fc fusion was shown to stain cells expressing VCAM-1 on their surface by FACs analysis. This approach for expressing soluble alpha4beta1 should be generally applicable to a range of integrins.

Expression, characterization and purification of simian immunodeficiency virus soluble, oligomerized gp160 from mammalian cells.

The envelope glycoprotein, gp160, of human (HIV) and simian (SIV) immunodeficiency viruses mediates virus-host cell binding followed by fusion of the viral and plasma membranes. The envelope proteins are known to exist as non-covalently associated oligomers on the virus surface. The production of permanent mammalian cell lines that constitutively secrete relatively high levels of soluble forms of SIV gp160 is described and we show that these proteins are secreted predominantly as tetramers with lower levels of dimer forms. Oligomeric forms were purified to greater than 90% purity using a simple gel filtration method. The purified proteins bind CD4 suggesting that they remain in their native conformation. The purified oligomeric proteins provide the basis for more relevant structural, functional and immunological studies than recombinant gp120 as they more closely resemble the envelope protein oligomer.

The pyruvate dehydrogenase complex of Escherichia coli K12. Nucleotide sequence encoding the dihydrolipoamide acetyltransferase component.

The nucleotide sequence of the aceF gene, which encodes the dihydrolipoamide acetyltransferase component (E2) of the pyruvate dehydrogenase complex of Escherichia coli K12, has been determined using the dideoxy chain-termination method. The aceF gene comprises 1887 base pairs (629 codons excluding the initiation codon AUG); it is preceded by a short intercistronic segment of 14 base pairs containing a good ribosomal binding site, and it is followed closely by a potential rho-independent terminator. The results extend by 1980 base pairs the previously sequenced segment of 3780 base pairs containing the structural gene (aceE) of the pyruvate dehydrogenase component (E1) and they confirm that aceE and aceF are the proximal and distal genes of the ace operon. The amino terminus, carboxy-terminal sequence and amino acid composition of the acetyltransferase subunit predicted from the nucleotide sequence are in excellent agreement with previous studies with the purified protein. The predicted molecular weight (Mr = 65959) confirms experimental values derived from sedimentation equilibrium analysis and indicates that the higher values (78000-89000) that have been reported are due to unusual features of the protein that lead to anomalous mobilities during sodium dodecyl sulphate/polyacrylamide gel electrophoresis and in gel filtration. The primary structure fully supports conclusions, based on limited tryptic proteolysis, that the acetyltransferase subunit possesses two heterologous domains: the lipoyl domain and the subunit binding and catalytic domain. The lipoyl domain corresponds to the amino-terminal segment of the protein. It is acidic and contains three remarkably homologous repeating units of approximately 100 amino acids, each possessing a potential lipoyl binding site and a region that is characteristically rich in alanine and proline residues. The subunit binding and catalytic domain occupies most of the residual polypeptide in the carboxy-terminal segment.

The pyruvate dehydrogenase complex of Escherichia coli K12. Nucleotide sequence encoding the pyruvate dehydrogenase component.

The nucleotide sequence of a 3780-base-pair segment of DNA containing the aceE gene encoding the pyruvate dehydrogenase component (E1) of the pyruvate dehydrogenase complex of Escherichia coli, has been determined by the dideoxy chain-termination method. The aceE structural gene comprises 2655 base pairs (885 codons, excluding the initiation codon AUG), it is preceded by a good ribosome binding site and several potential RNA polymerase binding sites. Its polarity and location in the restriction map of the corresponding segment of DNA are consistent with it being the proximal gene in the ace operon, as defined in previous genetic and post-infection labelling studies. The relative molecular mass (99474), composition (885 amino acids), amino-terminal residue and carboxy-terminal sequence predicted from the nucleotide sequence are in excellent agreement with published information obtained from studies with the purified pyruvate dehydrogenase component (E1). The nucleotide sequence also contains a second gene (gene A) situated upstream of the aceE gene. It appears to be an independent gene containing 708 base pairs (236 codons) and encoding a weakly expressed product (protein A; Mr = 27049) of unknown function.

Nucleotide sequence of the lipoamide dehydrogenase gene of Escherichia coli K12.

The nucleotide sequence of a 1980-base-pair segment of DNA, containing the lpd gene encoding the lipoamide dehydrogenase component (E3) of the pyruvate dehydrogenase complex of Escherichia coli K12, has been determined by the dideoxy chain-termination method. The lpd structural gene comprises 1419 base pairs (473 codons, excluding the initiating AUG codon). It is preceded by a good promoter and an excellent ribosome binding site and it ends with a typical rho-independent terminator sequence. The results confirm that the lpd gene is an independent gene linked to, but not part of, the ace operon that encodes the E1 and E2 components of the pyruvate dehydrogenase complex. The location and transcriptional polarity of the lpd gene relative to the restriction map of the corresponding region of DNA, are completely consistent with previous genetic and post-infection labelling studies. The composition, Mr (50554 or 51274 if the FAD cofactor is included), amino-terminal sequence and carboxy-terminal sequence predicted from the nucleotide sequence are in excellent agreement with previous studies on the purified enzyme. The enzyme also exhibits a remarkable degree of sequence homology with peptides of the pig heart enzyme and with other pyridine nucleotide disulphide oxidoreductases whose sequences have been defined: human erythrocyte glutathione reductase and plasmid-encoded mercuric reductase.