SEARCH · Search PubMed
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The properties and sequence of an oligomeric antigen of Treponema pallidum are presented. Antigen C1-5 assembles into oligomers of 140,000 and greater. The nucleotide sequence predicts an open reading frame for a protein monomer of 19,400, confirmed by amino-terminal sequencing of the recombinant antigen.
The growth of Treponema denticola in a complex medium was stimulated by human ceruloplasmin and, to a lesser extent, albumin, alpha-1-antitrypsin, alpha-1-acid glycoprotein, and alpha-2-macroglobulin. Human ceruloplasmin could stimulate growth of T. denticola as well as or better than 5% rabbit serum, with a dose-response relationship of between 1 and 5 mg of ceruloplasmin per ml of culture medium. These results suggest that ceruloplasmin could substitute for rabbit serum in stimulating the growth of T. denticola.
Penicillin-binding proteins (PBPs) of Treponema pallidum subsp. pallidum (T. pallidum) were characterized by using [3H]penicillin G and a conjugate consisting of ampicillin and 125I-labeled Bolton-Hunter reagent. Both antibiotics specifically radiolabeled proteins with molecular masses of 94, 80, 63, and 58 kilodaltons (kDa); 125I-labeled Bolton-Hunter reagent-ampicillin also radiolabeled several polypeptides with lower molecular masses. The 94- and 58-kDa proteins demonstrated the highest binding affinities for [3H]penicillin G and were radiolabeled at concentrations of 8 and 40 nM, respectively. Radiolabeling of PBPs was detectable after 1 min of incubation in 1 microM [3H]penicillin G and was nearly maximal within 10 min. The rapidity of penicillin binding contrasted with the observation that only 40% of virulent treponemes became immobilized during prolonged incubation in vitro with a much higher concentration (1 mM) of unlabeled penicillin. Two lines of evidence indicated that most, if not all, of the PBPs are integral cytoplasmic membrane proteins: (i) preincubation of organisms in 0.1% Triton X-100 solubilized nearly all of the outer membranes but did not affect radiolabeling of PBPs, and (ii) except for the 80-kDa protein, the PBPs partitioned into the detergent phase following extraction with the nonionic detergent Triton X-114. The presence of peptidoglycan in T. pallidum was confirmed by the detection of muramic acid in the sodium dodecyl sulfate-insoluble, proteinase K-resistant residue obtained from Triton X-114-extracted organisms.
A library of Treponema pallidum DNA was established in the direct selection vector pUN121. Six clones carrying a gene coding for a 33-kilodalton T. pallidum flagellum subunit were identified by colony hybridization with an oligodeoxynucleotide probe based on the N-terminal amino acid sequence of this subunit. An open reading frame of 286 amino acids with the expected N-terminal sequence and absence of cysteine residues was identified. The deduced protein had a calculated molecular weight of 31.3 kilodaltons. We propose to name this flagellar subunit FlaB2. FlaB2 shows a significant amino acid homology with flagellins of several remotely related bacterial species. This homology was most pronounced corresponding to the C-terminal and N-terminal parts of the protein, whereas the central region was variable.
Human oral spirochetes are prominent inhabitants of subgingival plaque in patients with periodontal disease. By immunoelectron microscopy using protein A-gold complexes and either polyclonal mouse antiserum against the 53-kDa antigen or 53-kDa-antigen-specific monoclonal antibody, a major polypeptide antigen, with a molecular weight of 53,000 (molecular size, 53 kilodaltons [kDa]), of a human oral spirochete, Treponema denticola ATCC 33520, was found to localize on the surface of the outer envelope.
We report the construction of expression plasmids carrying two Treponema pallidum genes encoding for the 42-kilodalton membrane protein TmpA (treponemal membrane protein A) and the 34-kilodalton membrane protein TmpB. Using the leftward promoter of bacteriophage lambda, which is controlled by a thermosensitive repressor, we obtained a high level of heat-inducible synthesis of TmpA and TmpB in Escherichia coli K-12. Both proteins were purified to near homogeneity, and the presence of antibodies to TmpA and TmpB in human sera was determined by an enzyme-linked immunosorbent assay. Whereas in all 44 serum samples from untreated patients in the secondary and early latent stages of syphilis, high levels of anti-TmpA antibodies were detected, only 34 serum samples contained anti-TmpB antibodies. As has been previously observed for TmpA, a correlation was found between the presence of anti-TmpB antibodies and anti-cardiolipin antibodies, suggesting that the level of antibodies to TmpB drops soon after successful antibiotic treatment. We concluded that, in contrast to TmpA, TmpB is not suitable for serodiagnostic purposes as a single antigen, because a significant fraction of sera from syphilitic patients was nonreactive with TmpB.
A number of the major pathogen-specific immunogens of Treponema pallidum were characterized recently as amphiphilic, integral membrane proteins by phase partitioning with Triton X-114 (J. D. Radolf, N. R. Chamberlain, A. Clausell, and M. V. Norgard. Infect. Immun. 56:490-498, 1988). In the present study, we demonstrated that the same membrane immunogens (designated as detergent phase proteins [DPPs]) become radiolabeled upon in vitro incubation of T. pallidum with various 3H-labeled fatty acids. Radioimmunoprecipitation with a monoclonal antibody confirmed that the 3H-labeled 47-kilodalton protein corresponded to the well-characterized treponemal antigen with the identical apparent molecular mass. Failure to detect 3H-labeled DPPs following incubation with erythromycin confirmed that protein acylation required de novo protein synthesis by the bacteria. When treponemes were incubated with [3H]myristate, [3H]palmitate, or [3H]oleate, radiolabeled proteins corresponding to the DPPs were detected upon autoradiography. Demonstration that a number of the abundant membrane immunogens of T. pallidum are proteolipids provides information to help clarify their membrane association(s) and may serve to explain their extraordinary immunogenicity.
The 47-kilodalton (kDa) major integral membrane immunogen of Treponema pallidum was recently found to be a proteolipid. Similar two-dimensional electrophoretic mobilities and common hydrophobic properties displayed by the native (T. pallidum) and recombinant (Escherichia coli) 47-kDa antigens suggested that the recombinant antigen also possesses covalently bound lipid. Both intact E. coli and E. coli minicells acylated the 47-kDa antigen; immunoprecipitation with a monoclonal antibody specific for the 47-kDa immunogen supported the contention that the acylated product of E. coli corresponds to the cloned T. pallidum antigen. Triton X-114 phase partitioning was used to compare the relative hydrophobicities of 47-kDa molecules synthesized by in vitro and in vivo protein translation systems. The products synthesized by T. pallidum, intact E. coli, or E. coli minicells were hydrophobic, while the protein synthesized in an E. coli cell-free translation system was hydrophilic. Processing experiments with E. coli suggested that the primary gene translation product of the protein is not synthesized in a precursor form, unlike other bacterial proteolipids. These results indicate that the hydrophobicity of the 47-kDa integral membrane protein is conferred substantially by the covalently attached lipid(s). The biochemical similarities between the native and recombinant 47-kDa proteolipids will provide a foundation for future investigations into the structure and immunogenicity of this integral membrane protein of T. pallidum.
The experimental infection of 25 young chicks with Treponema hyodysenteriae was carried out. Treponemes were isolated from 21 of 25 chicks on day 21 after inoculation. The ceca of chicks infected with treponemes were atrophied. The lumen was filled with a white watery fluid instead of digested feed. In some infected chicks, a cecal core was observed with the fluid in the cecum. The cecal core was grayish, hard, and rod shaped. It consisted of eroded cells and debris of treponemes and resembled the pseudomembrane. Bloody mucus was also observed in one chick. The thickness of the mucosae in 17 of 25 chicks were markedly increased. The histological changes were classified into two types. In the case of regressive changes of epithelial cells which mean severe erosion, the laminae propriae were exposed. Hemorrhage, edema, and heterophil infiltration in the laminae propriae were also confirmed. Numerous treponemes were observed within the edematous area under the remaining epithelia and also invaded the epithelial cells and laminae propriae. In the other case, progressive changes, that is, hyperplasia of mucosal epithelial cells and elongation of the crypt, were observed. The epithelia consisted mainly of cuboidal basophilic cells, mitotic cells, and goblet cells. The mitotic cells increased in number and were also observed near the superficial luminal surface of the ceca. Mucous goblet cells were also considerably increased in number. The erosion of superficial luminal epithelial cells was not so severe, but edema in laminae propriae was frequently observed. Electron-microscopic observation demonstrated that the basophilic epithelial cells were polyribosome rich, mitochondria poor, and lipid droplet poor. Furthermore, tonofibril-like structures under the terminal web in cytoplasms were lost, and numerous membrane-bound vesicles at the terminal web with free ribosomes were observed. In places, a number of vesicles were observed between microvilli, and some vesicles were released from the apical cytoplasm of epithelial cells. Microvilli also became scarce and irregular. The junctional complexes between cuboidal cells became unclear, the interdigitations between cuboidal cells became loose, and the intercellular spaces were widened. In conclusion, the appearance of numerous membrane-bound vesicles at the terminal web and the dilated intercellular space indicates increased secretion of water and electrolytes, and the abnormal proliferation of such immature epithelial cells indicates impaired absorption. On the basis of our findings, we can say that diarrhea induced by T. hyodysenteriae can be developed by synergism between impaired absorption and increased secretion by intestinal mucosae with hyperplastic immature epithelial cells.(ABSTRACT TRUNCATED AT 400 WORDS)
The adherence of Treponema denticola to ligands on cell surfaces or in basement membranes of periodontal tissues might play an important role in its pathogenicity. A direct microscopic assay was used to examine the binding of T. denticola to fibronectin and other protein substrates adsorbed on plastic cover slips. All strains of T. denticola that were tested adhered to fibronectin but to different degrees. The strains which bound in high numbers frequently bound by their tips. Type strain ATCC 33520 bound to fibronectin in high numbers (149 +/- 11.3 bacteria per microscopic field), with 60% bound by the tips. Strain e' bound in high numbers (140 +/- 10.2) and had the highest percentage of tip binding (98%); strain e bound in lowest numbers (39 +/- 8.2) and had the lowest percentage of tip binding (15%). Laminin supported binding at a level similar to that of fibronectin, as did fibronectin fragments which contained the cell binding domain peptides, RGDS. Type IV collagen and non-RGDS peptides did not support binding. Binding to fibronectin and laminin was inhibited by the addition of antifibronectin and antilaminin antibodies. By lowering the incubation temperature from 37 to 4 degrees C, the number of cells that attached decreased by 60% and tip binding was reduced by 50%. Pretreatment of the cells with collagen did not affect binding, whereas fibronectin pretreatment enhanced binding by 50% and laminin pretreatment resulted in a decrease of 60%. T. denticola adheres by its tips to fibronectin-coated surfaces, which suggests that fibronectin-specific adhesins cluster at the tips.
The chymotrypsinlike protease gene (prtA) from Treponema denticola ATCC 35405 was isolated from a lambda gt11 clone bank as one of several clones expressing protease activity. The DNA from one positive clone capable of hydrolyzing type IV collagen was subcloned into plasmid vector pUC119 for further analysis. Deletion analysis of subclone pXQ27.2 revealed the approximate location of the prtA gene on the DNA insert. DEAE-Sephadex chromatography of crude cell extracts of the subclone revealed two distinct T. denticola enzymes, one hydrolyzing SAAPNA (succinyl-L-alanyl-L-alanyl-L-prolyl-L-phenylalanine-p-nitroanilide [chymotrypsin substrate]) and the other hydrolyzing PZ-PLGPA (phenylazobenzyl-oxycarbonyl-L-leucylglycyl-L-prolyl-D -arginine [collagenase substrate]). Each activity was purified to near homogeneity and exhibited by sodium dodecyl sulfate-polyacrylamide gel electrophoresis estimated molecular sizes of 67 and 36 kDa, respectively. Western blot (immunoblot) analysis demonstrated that only the 67-kDa SAAPNA-hydrolyzing enzyme reacted with antibody against the T. denticola chymotrypsinlike protease. The purified SAAPNA-hydrolyzing enzyme degraded type IV collagen, laminin, and fibronectin, but not type I collagen. These results indicate that the prtA gene coding for the chymotrypsinlike protease from T. denticola has been isolated. Another distinct gene encoding an enzyme hydrolyzing PZ-PLGPA appears to be adjacent to the prtA gene.
A number of immunological methods were used to localize a cell-associated Treponema denticola chymotrypsinlike protease. Indirect immunofluorescence staining, immunogold labeling, and an enzyme-linked immunosorbent assay all indicated that the protease was attached to the outside of the cell envelope. The invasive capability of T. denticola was evaluated by following the degradation of a reconstituted basement membrane material (Matrigel) and the release of spirochetes from the gel. Under conditions where the chymotrypsinlike activity was increased, more spirochetes migrated from the gel. Protease inhibitors strongly reduced the number of cells that moved out of the gel. The purified chymotrypsinlike protease degraded the basement membrane components type IV collagen, laminin, and fibronectin. The study suggests that the T. denticola chymotrypsinlike protease may play an important role in the invasion and destruction of basement membrane.
Treponema pallidum subsp. pallidum and Escherichia coli incorporated exogenous [3H]palmitate into the 34-kilodalton (kDa) pathogen-specific antigen of T. pallidum. Radiolabeled fatty acid remained associated with the protein upon immunoprecipitation and after boiling in sodium dodecyl sulfate, acetone precipitation, and extensive extractions in organic solvents, suggesting that the fatty acid was covalently bound to the protein. Detection of [3H]palmitate after alkaline and acid hydrolyses confirmed the identity of the incorporated label. Globomycin inhibited maturation of the recombinant 34-kDa antigen, suggesting that E. coli uses the lipoprotein-specific signal peptidase II to process the treponemal antigen. Globomycin also inhibited processing of the 34-kDa antigen, as well as the 44.5- and 15-kDa antigens, in T. pallidum, implying that T. pallidum also possesses the lipoprotein export pathway common to both gram-negative and gram-positive bacteria. Ethanol inhibited processing of the 34-kDa antigen in minicells, suggesting that the 34-kDa antigen normally is translocated through the cytoplasmic membrane. Comparison of the Triton X-114 phase partitioning behavior of the 34-kDa antigen produced either by minicells or by a cell-free translation system indicated that the covalent attachment of fatty acid conferred hydrophobic biochemical properties to the 34-kDa antigen, consistent with the hypothesis that the attached lipid anchors the 34-kDa antigen into the membrane.
Macrophages were found to phagocytize Treponema pallidum subsp. pallidum attached to polycarbonate filters. This environment simulated the in vivo interaction of surface-adherent treponemes with macrophages. The phagocytosis of T. pallidum subsp. pallidum was found to proceed slowly. Heat-killed T. pallidum subsp. pallidum were susceptible to opsonization with 2% immune serum, whereas live treponemes were resistant to this concentration of antibody. High concentrations of immune serum were found to increase phagocytosis of the spirochetes. Live T. pallidum subsp. pallidum had bound limited quantities of immunoglobulin G in vivo, and only opsonization with 20% immune serum resulted in a detectable increase in surface-bound immunoglobulin in vitro. Kinetic studies suggested a steady rate of phagocytosis that is considerably slower than with other bacteria. Scanning electron microscopy studies of the phagocytizing macrophages showed that the treponemes were detached from the membrane filters and scooped onto the ruffled portion of the macrophage surface. This lengthy physical process, along with the lack of a dramatic increase in ingestion after opsonization, may account for the slow rate of phagocytosis.
The role of cell-mediated immunity against infection with Treponema pallidum subsp. pertenue in humans or experimental animals is unclear. Hamsters injected subcutaneously in the hind paws with 4 x 10(6) unfractionated lymph node cells or enriched lymph node T cells (immunoglobulin negative, Ia negative) from T. pallidum subsp. pertenue-immune hamsters were resistant to challenge with T. pallidum subsp. pertenue. The popliteal lymph nodes of hamsters that received immune cells weighed less and had significantly fewer treponemes than did lymph nodes from hamsters infused with cells from nonimmune donors. Furthermore, recipients of immune T cells failed to develop antitreponemal antibodies 21 days after challenge. Enriched T cells were obtained by flow cytometric separation by using monoclonal anti-Ia antibody 14-4-4s, which identified hamster B cells. Flow cytometric analysis by two-color immunofluorescent staining with anti-hamster-immunoglobulin and monoclonal anti-Ia antibody 14-4-4s confirmed that monoclonal anti-Ia antibody 14-4-4s recognized B cells. In addition, lymph node cells obtained after treatment with anti-Ia monoclonal antibody 14-4-4s and complement were 97% T cells, as determined by monoclonal antibody 20, a hamster T-cell marker. These results demonstrated that highly enriched T cells (immunoglobulin negative, Ia negative) from T. pallidum subsp. pertenue-immune hamsters conferred partial protection on hamsters against infection with T. pallidum subsp. pertenue.
Two structural endoflagellar genes of Treponema pallidum that encode the 34.5- and 31.0-kilodalton (kDa) polypeptides as detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis were cloned, sequenced, and expressed. We designated these genes flaB1 and flaB3. A DNA sequence analysis of flaB1 and flaB3 showed that each gene possesses a single open reading frame that encodes a polypeptide; these polypeptides have molecular masses of 31.1 and 31.0 kDa, respectively. Shine-Dalgarno ribosome-binding sequences were identified upstream from the initiation codons of each gene. In addition, a single consensus promoter sequence was identified 121 base pairs upstream from the initiation codon of flaB1, suggesting polycistronic transcription of flaB1 and flaB3. Computer-induced alignment showed that the FlaB1 amino acid sequence was identical at 206 positions (72%) to the FlaB3 sequence. Both genes were subcloned into pATH vectors and were expressed under the control of the trpE promoter. The expression products of flaB1 and flaB3 revealed fusion proteins having molecular masses of 61.0 and 59.0 kDa, respectively, which were identified on immunoblots by using specific anti-T. pallidum endoflagellar serum.
The adherence of Treponema denticola GM-1, TD-4, and MS25 to human gingival fibroblasts (HGFs) was studied to serve as an introduction to investigations into the interactions of these oral bacteria with human host cells. Under both aerobic (5% CO2) and anaerobic (85% N2 plus 10% H2 plus 5% CO2) environments, the interactions with the HGFs were such that strains GM-1 and MS25 were consistently more adherent than strain TD-4. Polyclonal antibodies to GM-1 inhibited GM-1 adherence by 70%, while MS25 and TD-4 showed differing degrees of cross-reactive inhibition, indicative of common but not identical epitopes on the surface of the three T. denticola strains. Pretreatment of the three strains with trypsin did not inhibit adherence; proteinase K did, however, inhibit this interaction by 80%. Trypsin pretreatment of the HGFs resulted in increases in adherence of 50 and 86% for GM-1 and MS25, respectively, while a decrease of 41% was noted for TD-4. Exposure of the T. denticola strains to sugars and lectin pretreatment of the HGFs implicated adherence mediation by mannose and galactose residues on the HGF surface. Periodate treatment of HGFs resulted in a 50% drop in adherence for GM-1 and MS25, but did not decrease that of TD-4. Addition of fetal bovine serum inhibited adherence of the three strains to differing degrees, with TD-4 being the most susceptible. Addition of purified fibronectin (100 micrograms/ml) resulted in greater than 50% inhibition in GM-1 and MS25 adherence, while a 25% increase occurred with TD-4. While strain differences were noted in some of the parameters studied, the results indicate two possibilities for T. denticola-HGF adherence: a lectinlike adhesin(s) on the T. denticola surface with affinity for galactose and mannose on the HGF surface, and a serum host factor(s) bridging T. denticola and HGFs.
We previously reported the complete primary structure of the 37-kilodalton endoflagellar sheath protein (FlaA) of Treponema pallidum. However, we were unable to determine the nucleotide sequence of flaA upstream of amino acid 10. The desired nucleotide sequence was obtained by use of a strategy based upon the polymerase chain reaction and was found to contain a consensus Escherichia coli promoter, a ribosomal binding site, and a 20-amino-acid signal peptide. Expression of FlaA in E. coli was achieved by cloning polymerase chain reaction-derived constructs lacking the native T. pallidum promoter into a temperature-inducible T7 expression system. Pulse-chase and ethanol inhibition analyses of protein processing in E. coli cells and minicells, respectively, indicated that processing of the FlaA precursor was incomplete. Native and recombinant FlaA were identical as assessed by antibody reactivity and sodium dodecyl sulfate- and two-dimensional polyacrylamide gel electrophoretic mobilities. Soluble FlaA was not detected in either the cytoplasmic or the periplasmic fractions of E. coli transformants. Fractionation of E. coli cell envelopes unexpectedly revealed that FlaA precursor and FlaA were associated with both the cytoplasmic and outer membranes. This is the first report of expression in E. coli of a T. pallidum protein which could not be cloned or expressed with its native promoter. Our data also indicate that information obtained in E. coli regarding the subcellular location of cloned treponemal proteins must be cautiously extrapolated to T. pallidum.