Generation and functional characterization of recombinant Porphyromonas gingivalis W83 FimA.

Porphyromonas gingivalis (P. gingivalis) is regarded as a keystone pathogen in destructive periodontal diseases. It expresses a variety of virulence factors, amongst them fimbriae that are involved in colonization, invasion, establishment and persistence of the bacteria inside the host cells. The fimbriae also were demonstrated to affect the host immune-response mechanisms. The major fimbriae are able to bind specifically to different host cells, amongst them peripheral blood monocytes. The interaction of these cells with fimbriae induces release of cytokines such as interleukin-1 (IL-1), IL-6, and tumor necrosis factor-α (TNF-α). The aim of this study was to generate recombinant major FimA protein from P. gingivalis W83 fimbriae and to prove its biological activity. FimA of P. gingivalis W83 was amplified from chromosomal DNA, cloned in a vector and transferred into Listeria innocua. (L. innocua).The expressed protein was harvested and purified using FPLC via a His trap HP column. The identity and purity was demonstrated by gel-electrophoresis and mass-spectrometry. The biological activity was assessed by stimulation of human oral epithelial cells and peripheral blood monocytes with the protein and afterwards cytokines in the supernatants were quantified by enzyme linked immunosorbent assay (ELISA) and cytometric bead array. Recombinant FimA could successfully be generated and purified. Gel-electrophoresis and mass-spectrometry confirmed that the detected sequences are identical with FimA. Stimulation of human monocytes induced the release of high concentrations of IL-1ß, IL-6, IL-10 and TNF-α by these cells. In conclusion, a recombinant FimA protein was established and its biological activity was proven. This protein may serve as a promising agent for further investigation of its role in periodontitis and possible new therapeutic approaches.

Production of biologically active light chain of tetanus toxin in Escherichia coli. Evidence for the importance of the C-terminal 16 amino acids for full biological activity.

The activity of the light (L) chain of tetanus toxin, and of mutants constructed by site-directed mutagenesis, was studied by expression and purification of the proteins from E. coli. Wild-type recombinant L chain (pTet87) was active in the inhibition of exocytosis from cultured bovine adrenal chromaffin cells, although at a level 5-15% of that of L chain purified from tetanus toxin. L chain mutants which terminated at Leu-438 (pTet89), or which contained a Cys-to-Ser mutation at residue 439 (pTet88) were equally as active as the full-length recombinant protein. The reduced activity of pTet87 L chain correlated with C-terminal proteolysis of the protein upon purification. A tryptic fragment derived from native light chain and which terminated at Leu-434 also showed reduced activity in the exocytosis assay, consistent with a requirement of the C-terminal region of the L chain for maximal activity. pTet87 L chain, but neither of the mutants, could be associated with purified H (heavy) chain to form a covalent dimer which induced the symptoms of tetanus in mice. The ability to form biologically active toxin using recombinant L chain will be of great value in structure-function studies of tetanus toxin.

Limited proteolysis of single-chain tetanus toxin by tissue enzymes, in cultured brain tissue and during retrograde axonal to the spinal cord.

Single-chain toxin was investigated in vitro and in vivo for limited proteolysis into the fully active two-chain toxin. Plasmin from serum, elastase and gelatinase from leucocytes, as well as clostripain from C. histolyticum cleaved single-chain toxin and increased by that way its ability to inhibit [3H]noradrenaline release in vitro. Cultured mouse brain generated fragments from 125I-single-chain toxin which were cell-associated. Some of them comigrated in electrophoresis with light and heavy chain after mercaptolysis. When injected i.v. into rats, 125I-single-chain-toxin disappeared from the blood with a half-life of about 11 h without signs of nicking. However, after its injection into the triceps surae muscle both single- and two-chain toxin were found in the ipsilateral ventral horn of the spinal cord. Thus single-chain toxin is subjected to limited proteolysis by enzymes involved in tissue damage, by cultured brain tissue, and during or after its retrograde axonal transport to the spinal cord. Limited proteolysis is necessary for the release of the light chain known to mediate the action of toxin on several systems.

Palytoxin promotes potassium outflow from erythrocytes, HeLa and bovine adrenomedullary cells through its interaction with Na+, K+ -ATPase.

Erythrocytes from four mammalian species were compared with regard to K+ loss triggered by palytoxin, to Na+, K+ -ATPase activity, and to ouabain sensitivity of both events. Palytoxin sensitivity (EC50) decreased in the order rat, man (approximately equal to 1 pM) greater than cattle (approximately equal to 500 pM) greater than dog (greater than 10 nM). Na+, K+ -ATPase activity, as measured by Rb uptake, was in the series rat greater than man greater than cattle greater than dog. The glycoside potently inhibited both palytoxin action and ATPase activity in man, cattle and dog erythrocytes, but weakly in those from rats. Ca2+ promoted the palytoxin effects on all erythrocytes. As shown for human erythrocytes, Sr2+ and Ba2+ but not Mg2+ can substitute for Ca2+, and sucrose can substitute for sodium chloride. Human HeLa and bovine adrenomedullary cells also lost their K+ within a few min when exposed to palytoxin (1-10 pM). Ouabain acted as a palytoxin antagonist on both cell types. We conclude that: (a) the ouabain binding site of Na+, K+ -ATPase is part of the palytoxin receptor in every cell type tested, (b) high palytoxin sensitivity is not necessarily accompanied by high ouabain sensitivity, and (c) active ion transport is not a precondition for the action of palytoxin or for its inhibition by ouabain.

Increase of permeability of synaptosomes and liposomes by the heavy chain of tetanus toxin.

In search of a role for the heavy chain of tetanus toxin in poisoning, its actions on natural and artificial membranes have been assessed. The heavy chain increases the permeability of synaptosomes to lactate dehydrogenase and potassium ions, and promotes the outward shift of the lipophilic cation tetraphenylphosphonium which is a particularly sensitive indicator for depolarization. Independent of the assay system the potency of the heavy chain is high, i.e. in the range of about 1 nM, whereas its efficacy is low. Its potency is decreased by the addition of the light chain and by treatment of the synaptosomes with the C-terminal fragment C of the heavy chain, but not with its N-terminal fragment beta 2. Single- or two-chain toxin itself is inactive, and so are the light chain or the two heavy chain fragments beta 2 and C. Liposomes were made from phosphatidylcholine and phosphatidylserine or gangliosides and loaded with calcein. At pH 6 the outflow of calcein is promoted in the order heavy chain greater than toxin much greater than fragment beta 2, and the action of toxin is promoted by ganglioside. At pH 5, fragment beta 2 is nearly as active as the heavy chain and more potent than the toxin. The heavy chain, but neither of the fragments, is strongly adsorbed in hydrophobic interaction chromatography and caused aggregation of polystyrene-divinylbenzene beads. Evidence for polymerization of heavy chains is lacking in zonal centrifugation. It is concluded that both domains of the heavy chain co-operate to exert the membranal events described, and that the heavy chain is partially hidden by the light chain in the complete toxin molecule.

Tetanus toxin and botulinum A and C neurotoxins inhibit noradrenaline release from cultured mouse brain.

Primary nerve cell cultures from the brainstem of embryonic mice take up [3H]noradrenaline. Release can be evoked by high K+ or sea anemone toxin II and depends on Ca2+. The cultures allow neurochemical studies on the long-term actions of clostridial neurotoxins. Tetanus and botulinum A and C neurotoxins partially inhibit the absolute and fractional release evoked by high K+, as well as the fractional basal release. The detection limit for the toxins is below 5 pM. Total radioactivity is higher in the poisoned cultures, although the initial velocity of uptake is not measurably influenced by tetanus or botulinum A toxin. Pretreatment with neuraminidase prevents the effects of botulinum A toxin and diminishes those of botulinum C and tetanus toxins. Within 6 days, the cultures partially recover from tetanus toxin poisoning. Antitoxin prevents the actions of the toxin, but only slightly promotes recovery. The data indicate close pharmacological analogies between the clostridial neurotoxins.

Characterization of an exported protease from Shiga toxin-producing Escherichia coli.

The gene for a novel, high molecular weight protein secreted by Shiga toxin-producing Escherichia coli (STEC) has been cloned, sequenced and characterized with respect to its activity. This gene, designated pssA, is localized on the large plasmid that also harbours the STEC haemolysin operon. Sequencing of a region comprising 10630nt revealed that the sequences flanking the pssA gene are composed of several remnants of different insertion elements. The PssA protein is produced as a 142kDa precursor molecule that, after N- and C-terminal processing, is released into the culture supernatant as a mature polypeptide of approximately 104kDa. The primary sequence of PssA is highly related to a family of autonomously transported putative virulence factors from different Gram-negative pathogens, which includes the Tsh protein of an avian-pathogenic E. coli strain, the SepA protein from Shigella flexneri and the EspC protein from enteropathogenic E. coli. A common motif present in all four proteins is reminiscent of the catalytic centre of certain serine proteases. PssA (protease secreted by STEC) indeed shows serine protease activity in a casein-based assay and is moreover cytotoxic for Vero cells. This activity of PssA and probably of other proteins of the Tsh family may be of functional importance during infection of the mucosal cell layer by the bacterial pathogen.

Expression of the Listeria monocytogenes EGD inlA and inlB genes, whose products mediate bacterial entry into tissue culture cell lines, by PrfA-dependent and -independent mechanisms.

Internalization of Listeria monocytogenes into nonphagocytic cell lines in vitro requires the products of the inlAB locus (J.-L. Gaillard, P. Berche, C. Frehel, E. Gouin, and P. Cossart, Cell 65:1127-1141, 1991). By generating isogenic mutants with a chromosomal in-frame deletion in either inlA or inlB, we have identified InlA and InlB as surface-bound proteins of L. monocytogenes with molecular weights of 88,000 and 65,000, respectively. These results were obtained with monoclonal antibodies raised against either protein and corroborated by N-terminal end sequencing of InlA and InlB. By immunoblot analysis, the production of both polypeptides was found to be strongly dependent on growth temperature and, particularly for InlB, on the presence of the PrfA regulator protein. Expression of InlA was not strictly dependent on the presence of the PrfA regulator protein. Transcription analysis of the inlAB locus revealed that the inlA gene was transcribed by several promoters, of which only one is PrfA dependent. This PrfA-dependent inlA promoter, which contains two base substitutions within its putative PrfA DNA-binding palindrome, is responsible for transcription of both inlA and inlB genes. A hitherto unrecognized promoter located 51 bp upstream of the GTG start codon of the inlB gene was also detected. Hence, inlA and inlB are transcribed both individually and in an operon by PrfA-dependent and -independent mechanisms. Tissue culture invasion assays employing various epithelial cell lines demonstrated that both InlA and InlB are required for invasion. In vivo studies using the mouse infection model revealed that both internalin mutants were attenuated for virulence.

Tetanus toxin: primary structure, expression in E. coli, and homology with botulinum toxins.

A pool of synthetic oligonucleotides was used to identify the gene encoding tetanus toxin on a 75-kbp plasmid from a toxigenic non-sporulating strain of Clostridium tetani. The nucleotide sequence contained a single open reading frame coding for 1315 amino acids corresponding to a polypeptide with a mol. wt of 150,700. In the mature toxin molecule, proline (2) and serine (458) formed the N termini of the 52,288 mol. wt light chain and the 98,300 mol. wt heavy chain, respectively. Cysteine (467) was involved in the disulfide linkage between the two subchains. The amino acid sequences of the tetanus toxin revealed striking homologies with the partial amino acid sequences of botulinum toxins A, B, and E, indicating that the neurotoxins from C. tetani and C. botulinum are derived from a common ancestral gene. Overlapping peptides together covering the entire tetanus toxin molecule were synthesized in Escherichia coli and identified by monoclonal antibodies. The promoter of the toxin gene was localized in a region extending 322 bp upstream from the ATG codon and was shown to be functional in E. coli.

Interaction of Listeria monocytogenes with mouse dendritic cells.

In this study, the interaction of murine dendritic cells with Listeria monocytogenes was investigated. Dendritic cells are efficient antigen-presenting cells, play a key role in the immune response, and are capable of migrating over substantial distances between sites of infection and lymphoid tissues. L. monocytogenes EGD invaded dendritic cells, escaped from phagosomes into the cytoplasm, and there directed actin nucleation, polymerization, and polarization in a typical fashion, thereby achieving intracellular movement and cell-to-cell spread. The internalization process appears to be independent of the inl locus. Interestingly, an intact microtubular function was essential for efficient uptake, whereas in a previous report, microtubule disruption did not affect bacterial spread in Caco-2 cells. The results obtained also suggest that L. monocytogenes binds to glycosylated receptors of dendritic cells. Uptake of Listeria cells was mediated by a protein kinase-dependent transducing phosphorylation signal that induces the actin polymerization-dependent phagocytic process. To achieve efficient uptake, de novo protein synthesis of eukaryotic and prokaryotic cells is also required. Despite the killing of dendritic cells, wild-type bacteria were found to persist in small numbers in some cells for at least 24 h. When different isogenic mutants of the EGD strain were analyzed for their capability to interact with dendritic cells, it was observed that some virulence-attenuated mutants (i.e., prfA and delta hly) persisted in large numbers for even longer times. Invasion of dendritic cells by L. monocytogenes, which in turn could result in either cell death or persistent infection, might have an important role in the pathogenesis of listeriosis, leading to impaired immune responses with inefficient bacterial clearance and/or promoting bacterial spread.