Search

The outer membranes (OMs) from serovars a, b, and c of Treponema denticola, originally isolated from periodontal patients, were prepared. Dialysis of the OMs against 20 mM MgCl2 yielded the aggregable (A) and the nonaggregable (NA) moieties of the OMs. The absence of muramic acid, adenosine triphosphatase, hexokinase, and nucleic acid as well as electron microscopy indicated that the OM preparations were homogeneous. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the A and NA moieties of the OMs showed approximately 25 Coomassie brilliant blue R-250 stain-positive bands or 47 silver-stained polypeptides. The relative molecular masses ranged between 14 and 97 kDa. The electrophoretic polypeptide profiles of the A and NA moieties shared many similarities among serovars a, b, and c. However, they exhibited variation in the overall pattern, intensity, or location of the polypeptide stained zones. This was especially true for serovar b. Two-dimensional electrophoretic studies showed an excess of 100 silver-stained spots with isoelectric points of 4.6 to 7.0 and relative molecular masses in the 14- to 97-kDa range. The OMs contained simple proteins, glycoproteins, and lipoproteins. The NA moieties of the OMs contained 4 to 6, 10 to 12, and 4 to 6 glycopeptides as well as two, seven, and two lipoprotein bands for serovars a, b, and c, respectively. The A moieties of the OMs showed 7 to 9, 11 to 13 and 5 to 6 glycopeptides as well as four, five, and three lipoprotein bands for serovars a, b, and c, respectively. Lipopolysaccharide was detected in the OMs of the three serovars following removal of proteins with proteinase K, pronase and silver staining of sodium dodecyl sulfate-polyacrylamide gels, or removal of lipopolysaccharide from the OMs by hot phenol extraction. The 66- and 53-kDa bands were present in serovars b and c, while a band with a relative molecular mass of 45 kDa was present only in serovar c. Endotoxin-like activity was also shown in the OMs of the three serovars by the Limulus amebocyte clotting assay and the chick embryo lethality test. This is the first report on selected biochemical properties of the OM macromolecules of three known serovars of T. denticola.

Treponema denticola, Porphyromonas gingivalis, and Bacteroides forsythus each possesses an enzyme(s) that hydrolyzes the synthetic substrate benzoyl-DL-arginine-naphthylamide (BANA). The presence of these organisms in a subgingival plaque sample can be determined by the ability of the plaque to hydrolyze BANA. In the present study, we describe the usefulness of the BANA test at various stages of a clinical trial of the efficacy of metronidazole in the treatment of periodontal disease. A BANA-positive test was significantly associated with high levels and proportions of spirochetes in the plaque, so that it provided information comparable with that which could be obtained by a microscopic examination of the plaque. Patients with such anaerobic spirochetal infections were randomly assigned to a group receiving either metronidazole or placebo (250 mg, three times a day) for one week and whose teeth were scaled and root-planed. The advantages of the decision that metronidazole be used were apparent from the comparison with the results obtained in the patients who received only the scaling and root planing. The initially BANA-positive teeth in the patients treated with metronidazole, scaling, and root planing gained attachment and exhibited a significant reduction in the need for periodontal surgery, when compared with the BANA-positive teeth in the patients who received only placebo, scaling, and root planing. After the conclusion of this therapy, those teeth with persistent BANA-positive plaques had significantly higher proportions and levels of spirochetes than did the teeth with BANA-negative plaques.(ABSTRACT TRUNCATED AT 250 WORDS)

The protein composition of 18 clinical isolates of Treponema hyodysenteriae from pigs with swine dysentery in Australia were compared by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblot analysis. Coomassie Blue stained SDS-PAGE-profiles of whole cell and outer membrane (OM) proteins demonstrated the same gel pattern among the T. hyodysenteriae isolates, particularly the OM proteins in the molecular mass (Mr) range of 30 kDa to 40 kDa. The T. hyodysenteriae isolates were categorised into two distinct groups (A and B) based on the strain-variability in the 37 kDa OM protein. Immunoblotting of whole cell proteins after SDS-PAGE using serum from rabbits and pigs immunised with known T. hyodysenteriae serotypes revealed a number of common immunoreactive bands in all isolates. LPS typing of the T. hyodysenteriae isolates by immunoblotting with the rabbit antiserum revealed one additional serotype emphasising the LPS heterogeneity among the strains isolated from geographic locations in Australia, Great Britain and the U.S.A. Immunoblotting of the OM preparations revealed several common immunoreactive polypeptides corresponding to Mr values of 34 kDa to 30 kDa among the T. hyodysenteriae and T. innocens isolates but a distinct 39 kDa found only in the T. hyodysenteriae isolates. Trypsin proteolysis of intact. T. hyodysenteriae cells caused selective loss of these and other major abundant proteins identifying the location of the 39 kDa, 36 kDa, 34 kDa and 30 kDa proteins on the cell surface and suggesting a possible role of these proteins in the pathogenesis of swine dysentery.

The protein profiles of Treponema pallidum and Treponema pertenue, the causative agents of syphilis and yaws, respectively, were compared by one- and two-dimensional gel electrophoresis. One-dimensional gels showed essentially no differences in the protein patterns of these treponemes. On two-dimensional gels most radiolabeled protein species were shared; however, variations were noticed in several minor protein species. Antigenic comparison by radioimmunoprecipitation and Western blotting also demonstrated similarities between these spirochetes. However, lactoperoxidase-catalyzed iodination of T. pallidum and T. pertenue suggested differences in their surface proteins.

To investigate the development of the humoral immune response of mice to infection with Treponema pallidum, Balb/c and C57Bl mice were injected intradermally with 10 x 10(6) virulent organisms. Serum samples were taken from the mice at weekly intervals after infection and used in the electrophoretic transblotting technique to detect T pallidum and T phagedenis biotype Reiter antigens. The serum samples taken from infected mice at 21, 42, 84, and 126 days after infection recognised two to 15 T pallidum antigens, with a gradual but continual increase in the number of antigens recognised. The same antisera to T pallidum recognised five cross reactive T phagedenis biotype Reiter antigens. Mice injected with 10 x 10(6) heat killed T pallidum organisms failed to recognise T pallidum antigens, as shown by the blotting technique. When mice infected with T pallidum were given antibiotics, the development of the humoral response was interrupted. These experiments indicated that mice respond more slowly than rabbits to T pallidum, which may be because T pallidum is weakly immunogenic in mice.

The periplasmic flagella of many spirochetes contain multiple proteins. In this study, two-dimensional electrophoresis, Western blotting (immunoblotting), immunoperoxidase staining, and N-terminal amino acid sequence analysis were used to characterize the individual periplasmic flagellar proteins of Treponema pallidum subsp. pallidum (Nichols strain) and T. phagedenis Kazan 5. Purified T. pallidum periplasmic flagella contained six proteins (Mrs = 37,000, 34,500, 33,000, 30,000, 29,000, and 27,000), whereas T. phagedenis periplasmic flagella contained a major 39,000-Mr protein and a group of two major and two minor 33,000- to 34,000-Mr polypeptide species; 37,000- and 30,000-Mr proteins were also present in some T. phagedenis preparations. Immunoblotting with monospecific antisera and monoclonal antibodies and N-terminal sequence analysis indicated that the major periplasmic flagellar proteins were divided into two distinct classes, designated class A and class B. Class A proteins consisted of the 37-kilodalton (kDa) protein of T. pallidum and the 39-kDa polypeptide of T. phagedenis; class B included the T. pallidum 34.5-, 33-, and 30-kDa proteins and the four 33- and 34-kDa polypeptide species of T. phagedenis. The proteins within each class were immunologically cross-reactive and possessed similar N-terminal sequences (67 to 95% homology); no cross-reactivity or sequence homology was evident between the two classes. Anti-class A or anti-class B antibodies did not react with the 29- or 27-kDa polypeptides of T. pallidum or the 37- and 30-kDa T. phagedenis proteins, indicating that these proteins are antigenically unrelated to the class A and class B proteins. The lack of complete N-terminal sequence homology among the major periplasmic flagellar proteins of each organism indicates that they are most likely encoded by separate structural genes. Furthermore, the N-terminal sequences of T. phagedenis and T. pallidum periplasmic flagellar proteins are highly conserved, despite the genetic dissimilarity of these two species.

Genetic relationships among two strains of Treponema pallidum (Nichols and KKJ) and a strain of T. pertenue were determined by measuring the degree of deoxyribonucleic acid sequence homology. The results in indicated that these three virulent, noncultivable treponemes were genetically indistinguishable. Like T. pallidum (Nichols), T. pertenue (Gauthier) had no detectable deoxyribonucleic acid sequence homology with T. phagedenis (biotype Reiter), T. refringens (biotype Noguchi), or with salmon sperm.

Properties of the attachment of Treponema hyodysenteriae to Henle intestinal epithelial (HIE 407) cells were examined. The frequency of attachment depended on the motility and viability of the spirochetes. Rabbit hyperimmune and swine convalescent antisera inhibited attachment. Treatment of HIE cells with neuraminidase had no effect on attachment; however, treatment of spirochetes with the enzyme decreased adherence significantly (P = 0.01). Attachment was inhibited by N-acetylneuraminic acid, D-glucuronic acid, and fetuin. Adherence was increased following coincubation with N-acetylglucosamine or yeast mannan. Surface antigens of T hyodysenteriae, isolated by chemical extraction, competitively inhibited adherence. Concentrated T hyodysenteriae culture supernatant fractions inhibited adherence, but concentrated phosphate buffered-saline washings of the spirochete and concentrated uninoculated media did not inhibit adherence. Sialic acid was detected in unwashed T hyodysenteriae and spent culture supernatant fractions in higher concentrations than from washed spirochetes and uninoculated media. It was concluded that the binding adhesins on T hyodysenteriae for cultured HIE cells may contain sialic acid residues.

The lipids of Treponema hyodysenteriae B78, the etiologic agent of swine dysentery, comprised 16.4% of the cell dry weight, and consisted of 37.4% glycolipids, 28.6% phospholipids, and 34.0% neutral lipids. Monogalactosyldiacylglycerol, a major lipid in all Treponema except Treponema pallidum, comprised 80% of the glycolipids. An unidentified galactolipid less polar than monogalactosyldiacylglycerol was also detected. Phosphatidylglycerol (19.5% of the total lipids) was the major phospholipid. Phosphatidylcholine, characteristically the major phospholipid of treponemes, comprised 6.1% of the total lipids. Cardiolipin and lysophosphatidylcholine were minor components. The alk-1-enyl ether forms of both the phospholipids (plasmalogens) and glycolipids predominated. The alk-1-enyl ether forms of monogalactosyldiacylglycerol, the unidentified galactolipid, phosphatidylglycerol, cardiolipin, and phosphatidylcholine were 88.3, 96.4, 74.8, 60.6, and 6.3%, respectively. The acyl and alk-1-enyl chains of the organism were qualitatively similar and differed dramatically from those of the medium indicating a capability for fatty acid synthesis that most Treponema do not possess. Saturated C14, C15, and C16 chains comprised more than 95% of the acyl and alk-1-enyl groups. About 25% of the chains were iso-15:0, anteiso-15:0, and other branched moieties.

The technical variables of the solid-phase immunofiltration assay (SPIA) for the detection of antibodies bound to antigens on a solid-phase filter have been investigated. The binding to solid-phase filters of 125I-labelled axial filament proteins derived from Treponema phagedenis and the optimal conditions for blocking non-specific protein binding were analysed. Axial filament was applied to nitrocellulose, Hybond Nylon and Zeta Probe. After extensive rinsing, the highest amount (68%) of axial filament was observed bound to Zeta Probe. However, blocking non-specific protein binding by pre-wetting the filter with rinsing buffer containing 0.5% Tween 20, prevented the binding of protein to the filter only when nitrocellulose was used as solid phase. Tween 20 (0.5%) in the rinsing and incubation solutions was found to be necessary for the reduction of non-specific binding of contaminants in turbid sera. However, the use of such solutions resulted in a substantial leakage of antigen (47%) during rinsing procedures. Binding of antigen-specific antibody was analysed using 125I-labelled protein A. The maximal possible binding of the antibody occurred within 5 min when the antibody solution was filtered. For optimal binding of 125I-labelled protein A an incubation time of 1 h was needed. It is suggested that solid-phase immunofiltration may provide a rapid alternative for radioimmunoassays or enzyme immunoassays for the detection of specific antibodies.

The Treponema denticola content of plaque was quantitatively estimated for samples taken from periodontitis patients as well as periodontally healthy subjects among two separate human populations. The populations studied included military volunteers and civilians at a university dental clinic. The plaque samples from each population were grouped according to pocket depth measurements at the collection site. A biotin-avidin enzyme-linked immunosorbent assay procedure was developed with a monoclonal antibody specific for a serovariety of T. denticola. T. denticola was present at significantly elevated levels in plaque samples collected from deep-pocket sites of patients with severe periodontitis relative to the healthy controls and a group with moderate disease. The ratio of T. denticola content per milligram of plaque in the deep pocket groups to that of the other two groups was about 2:1 for both populations. This is the first quantitative evidence of a positive relationship between a specific spirochete species and severe periodontitis.

"Treponema phagedenis" periplasmic flagella (PF) have two major protein bands at molecular weights of 33,000 and 39,800 as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (R. J. Limberger and N. W. Charon, J. Bacteriol. 166:105-112, 1986). By use of Western blotting and a polyclonal antiserum directed toward the 33,000-molecular-weight PF protein, cell lysates of 12 species of spirochetes were surveyed for reactivity. Eight species of Treponema as well as Spirochaeta aurantia were positive. The results suggest that epitopes residing on the 33,000-molecular-weight PF protein of "T. phagedenis" are evolutionarily well conserved among the spirochetes.

Treponema phagedenis is an anaerobic, motile spirochete with several periplasmic flagella (PFs) at each cell end. This study provides the first genetic evidence that multiple protein species are associated with the PFs. In addition, these proteins were found to reside together on a given PF. Nonmotile mutants which lacked the PFs were isolated, and spontaneous revertants to motility regained the PFs. These results suggest that the PFs are involved in the motility of T. phagedenis. Isolated PFs had two major protein bands with molecular weights of 33,000 and 39,800, as revealed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Western blots with monoclonal and polyclonal antibodies indicated that both proteins were absent in the PF mutants but present in the revertants. Immunoelectron microscopy revealed that the 39,800-molecular-weight protein was distributed along the entire PF. Immunoprecipitation analysis suggested that the 39,800- and 33,000-molecular-weight proteins were closely associated in situ.

Microscopic agglutination (MA) analysis and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) were used to determine strain and species similarities and dissimilarities among three species of oral anaerobic spirochetes, Treponema denticola, Treponema pectinovorum, and Treponema vincentii. The MA analysis revealed a diversity of serologic reactivity or sharing of common antigens within each species. However, there was no cross-reactivity or sharing of common antigens among the three species. Distinct SDS-PAGE whole-cell electrophoretograms for each species were obtained. The banding patterns for 16 T. denticola strains revealed 30 distinct proteins, while the banding patterns for 5 strains of T. pectinovorum and 2 strains of T. vincentii revealed 26 and 35 distinct proteins, respectively. Analysis of the electrophoretograms showed that their respective banding patterns could be used to distinguish the three species from one another. In addition, strain differences within each species could be detected. There was a correlation between MA analysis and SDS-PAGE analysis. It is thus suggested that both MA and SDS-PAGE analysis be included in classification schemes for the identification of oral spirochetes.

The outer sheath carrying a polygonal array was isolated from an oral treponeme, Treponema sp. strain E-21, by disruption of cells by means of repeated freeze-thawing and by removal of flagella under acidic conditions followed by linear sucrose density gradient centrifugation. Electron microscopy revealed that the outer sheath was isolated as a triple-layered vesicle having a polygonal array, free of flagella and wall membrane complex. Using optical diffraction, negatively stained preparations of the outer sheath fragments showed that the polygonal array appeared to be composed of a hexagonal pattern with a predominant spacing of about 16.3 nm. The isolated outer sheath contained 49.7% protein, 30.8% total lipid, and 11.0% carbohydrate. Phospholipid comprised about 95% of the total lipid. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed that the outer sheath was composed primarily of one major protein with an apparent molecular weight of about 62,000. The material from the isolated outer sheath solubilized with 1% sodium deoxycholate was reassembled into vesicles having a roughly polygonal array upon removal of the detergent by dialysis against 10 mM Tris-hydrochloride buffer with or without Mg2+.

Saturation reassociation assays with 125I-labeled treponemal DNAs show that Treponema hyodysenteriae is genetically unrelated to T. pallidum (Nichols), T. phagedenis biotype Reiter, and T. refringens biotype Noguchi. Pathogenic and nonpathogenic isolates of T. hyodysenteriae exhibited 28% sequence homology and had an extremely low guanine-plus-cytosine content (25.8%).

Biological activities of lipopolysaccharide-like (LPS-like, phenol/water extract) and endotoxin-like (butanol/water extract) preparations from Trepomena hyodysenteriae were examined. The treponemal phenol/water and butanol/water extracts were less toxic than E. coli LPS for murine peritoneal exudate cells (PECs). The treponemal phenol/water extract did not stimulate the production of interleukin-1 (IL-1) or tumor necrosis factor (TNF) from murine PECs. The treponemal butanol/water extract did induce production of IL-1 and TNF but at doses 5- to 50-fold higher than E. coli LPS. Natural killer cell activity was augmented by the treponemal butanol/water extract but not by the phenol/water extract. Suppression of a splenic anti-SRBC plaque forming cell response was observed when the LPS-like and endotoxin-like preparations from T. hyodysenteriae were administered 24 h prior to injection of the SRBC. These findings indicate that the butanol/water extracted material from T. hyodysenteriae is more biologically active than the phenol/water extracted material and that the treponemal endotoxin may contribute to the inflammatory response of swine dysentery by inducing IL-1 and TNF production.

The lipids of Treponema innocens, type strain B256, formerly considered a nonpathogenic isolate of T. hyodysenteriae, have been analyzed and compared with the lipids of T. hyodysenteriae. The lipids of T. innocens comprised 16% of the cell dry weight. Polar lipids amounted to about two-thirds of the total lipids and consisted of 61.9% phospholipids and 38.1% glycolipid. Neutral lipids consisted mainly of sterols. The phospholipids were principally phosphatidylglycerol, phosphatidylcholine, and cardiolipin. Minor amounts of lysophosphatidylcholine, sphingomoyelin, and a relatively nonpolar, unidentified phospholipid were present. The latter lipid has not been detected in T. hyodysenteriae. The glycolipid fraction of T. innocens contained a single component, monoglucosyldiglyceride, in contrast to the occurrence in T. hyodysenteriae of two components: monogallactosyldiglyceride and a less-polar glycolipid tentatively identified as acylmonogalactosyldiglyceride (the additional acyl moieties being 86.6% acetyl, 11.6% propionyl, and 1.6% n-butyryl groups). Alk-1-enyl ether analogs comprised 24.6% of the total phospholipids and glycolipid of T. innocens, or about one-third of the amount in T. hyodysenteriae. The acyl and alk-1-enyl moieties of T. innocens consisted of greater than or equal to 92% of 14:0, iso-15:0, and 16:0 chains. In contrast to T. hyodysenteriae, anteiso-15:0 moieties were not detected, and a reversed distribution of 14:0 and iso-15:0 alk-1-enyl moieties occurred in the two species.

A sodium deoxycholate-ethanol extractable antigen (DES-Ag) was obtained from four serotypes of Treponema denticola and characterized by chemical, physical, and serological procedures. The cross-reactivity of these antigens was demonstrated by indirect microhemagglutination, immunodiffusion, and immunoelectrophoresis with hyperimmune rabbit antiserum against each of the T. denticola serotypes. Antiserum against two nonoral treponemes, T. phagedenis biotype Reieter and T. pallidum Nichols strain, did not cross-react with the DES-Ag of T. denticola. Chemical analysis of the DES-Ag indicated that proteins (84%) and hexoses (12%) accounted for 96% of the total dry weight of the antigen. Trace amounts of N-acetylglucosamine (0.6%) were also detected. Fatty acids, including palmitic, oleic, and stearic, were identified by gas-liquid chromatography. Polyacrylamide gel electrophoresis of the DES-Ag of each serotype revealed the presence of two bands. The molecular weights of the bands were estimated to be 58,000 and 31,000 by comparing the electrophoretic mobility of the DES-Ag to that of five protein standards of known molecular weights. Although only a single precipitin line was observed by immunodiffusion when each antigen was reacted against its homologous antiserum, two precipitin lines were evident by immunoelectrophoresis. Antiserum against the DES-Ag of T. denticola was shown to agglutinate whole cells of the homologous serotype. Adsorption of this anti-DES-Ag serum with whole cells of T. denticola resulted in the elimination of precipitating antibodies to the DES-Ag by immunodiffusion. It is concluded that the DES-Ag is a component of the outer sheath of T. denticola.

A hemolysin produced by Treponema hyodysenteriae ATCC27164 was purified from broth filtrates by acetic and (NH4)2SO4 precipitations followed by ion exchange chromatography on diethylaminoethyl-Sephacel and gel filtration using Ultrogel AcA44. The purified hemolysin displayed only one band on polyacrylamide gel electrophoresis. By gel filtration the molecular weight was estimated as 74,000 daltons. The isolated hemolysin was oxygen resistant, heat labile and was not inactivated over a wide range of pH values. Further analysis indicated that this hemolysin was probably a polypeptide or a protein associated with lipids and nucleotides. Its action on rabbit erythrocytes which did not require any divalent cations could not be related to a lipolytic or proteolytic activity.

The sterol content of cellular lipids of Treponema hyodysenteriae, the agent of swine dysentery, was determined. When cultured in lipid-depleted brain heart infusion broth containing vesicles made from [4-14C]cholesterol and phosphatidylcholine, T. hyodysenteriae cells incorporated radioactive label. Most (95%) of this radioactivity was associated with bacterial membrane preparations. Lipids were extracted from radiolabeled cells and fractionated by silicic acid column chromatography. Components of the neutral lipid fraction were separated by reversed-phase high-performance liquid chromatography and were detected by monitoring both radioactivity and UV absorption (210 nm) of the column effluent. Cholesterol represented only about 5% of the total radioactivity in the bacterial neutral lipids. The remaining radioactivity was associated with a compound that did not absorb light at 210 nm. This lipid was purified and, on the basis of results from thin-layer chromatography and mass spectrometry, was identified as cholesterol (5 alpha-cholestan-3 beta-ol), a sterol lacking the unsaturated bond of cholesterol. Cholestanol was also present in cell-free culture broth, but only after growth of the spirochete. These results are evidence that cholesterol is used by T. hyodysenteriae for membrane synthesis. Cholesterol is converted to cholestanol in T. hyodysenteriae cultures and cholestanol is a major component (approximately 9% by weight) of T. hyodysenteriae cell lipids.

A lipopolysaccharide (LPS) was obtained from pathogenic Treponema hyodysenteriae by hot phenol-water extraction. Various effects of the LPS on host cells were examined in vitro. Toxicity for mouse peritoneal macrophages was observed after 10 h of incubation at concentrations as low as 15 micrograms of the LPS per ml. Marked enhancement of both complement (C3) and immunoglobulin G-Fc receptor-mediated internalization was noted in macrophages obtained from mice injected 6 days previously with 75 micrograms of the material. Incorporation of [3H]thymidine into murine splenocytes was elevated approximately fourfold when splenocytes were treated with 5 to 10 micrograms of LPS per ml. Incubation of the LPS with normal porcine serum resulted in the generation of a factor(s) that stimulated the migration of porcine leukocytes. Generation of the chemotactic activity was inhibited by heating the serum at 56 degrees C for 30 min before treatment with LPS. The results suggest that T. hyodysenteriae contains an LPS that is biologically active.

Lipooligosaccharides from Treponema hyodysenteriae serotypes 1 through 7, attenuated T. hyodysenteriae serotypes 1 and 2, and five strains of T. innocens were extracted with hot phenol water. The extracts were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis separation and analyzed by lipopolysaccharide selective silver staining and Western blot (immunoblot) immunodetection. Silver staining revealed the presence of two bands that ranged between 18,000 and 24,000 daltons and that were serotype specific for T. hyodysenteriae. Attenuation of pathogenic strains resulted in the loss of the higher-molecular-weight band. Four of five T. innocens strains also lacked this particular band. T. innocens 421 had six bands between 17,000 and 26,900 daltons. Western blots with hyperimmune rabbit sera and convalescent-phase swine sera revealed antigenic variation among serotypes of T. hyodysenteriae and attenuated serotypes of T. hyodysenteriae. Convalescent-phase swine sera failed to recognize lipopolysaccharides from T. innocens. Differences in results obtained by lipopolysaccharide selective silver staining versus immunoblotting of the lipopolysaccharide preparations probably indicate that these two methods identify separate characteristics of the same molecule.

Axial filaments of Treponema hyodysenteriae were purified and examined by transmission electron microscopy. The size and fine structure of the filaments were similar to those of Treponema zuelzerae and Treponema phagedenis biotype reiterii. Filaments were dissociated by heat and 2-mercaptoethanol and examined by polyacrylamide-gel electrophoresis. Six protein bands were evident, with approximate molecular weights of 39,000, 29,000, 27,000, 22,000, 21,000, and 18,500 daltons. All the proteins reacted with T hyodysenteriae and T innocens rabbit antisera, using the immunoblot technique.

An outer sheath was isolated from Treponema phagedenis biotype Reiter by our previously developed method (Masuda, K., and Kawata, T. 1982. J. Bacteriol. 150: 1405-1413). The isolated outer sheath was observed as a triple-layered, closed vesicle carrying a polygonal array by electron microscopy. The outer sheath was mainly composed of protein (41.0%), phospholipid (38.7%), and carbohydrate (11.0%). Sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of the isolated outer sheath in the presence of 2-mercaptoethanol (EtSH) gave one major protein band with an apparent molecular weight of about 69,000 and several minor protein bands. On the other hand, in the absence of EtSH, the major protein band disappeared but two new protein bands at positions of molecular weights of about 65,000 and 72,000 appeared. The SDS-PAGE profiles of the minor protein bands did not change with or without EtSH. Sodium deoxycholate (DOC)-solubilized materials from the isolated outer sheath were reassembled into thin membranous sheets carrying a roughly polygonal array upon removal of DOC by dialysis against Tris-HCl buffer in the absence of Mg2+.

The chemical composition of the outer envelope (OE) of Treponema phagedenis biovar Kazan 5 was investigated. After cultivation in a lipid-defined medium, the OE was removed from the cells with 0.7 mM sodium dodecyl sulfate. The solubilized OE was reaggregated by dialysis against 20 mM MgCl2, washed, lyophilized, and subjected to chemical analysis. The average yield of OE was 14.6% of the whole cell (WC) dry weight. The magnesium content was 0.683 mug/mg OE. Peptidoglycan components such as muramic acid and ornithine were detected in the WC but not in the OE, and diaminopimelic acid was absent in both WC and OE. The OE contained protein (60-73%), carbohydrate (1-2%), and lipid (4-5%), primarily polar lipid. The major polar lipids were monogalactosyldiglyceride (43%) and phospholipid (57%), of which phosphatidylcholine was the main phospholipid component, with phosphatidylethanolamine present in lesser amounts.

Purified axial filaments from eight serotypes of Treponema hyodysenteriae and two non-pathogenic intestinal spirochaetes were characterized by SDS-PAGE and Western blotting. Axial filaments of all ten strains had similar SDS-PAGE profiles; five major axial filament polypeptides were identified, with molecular masses of 43.8, 38, 34.8, 32.8 and 29.4 kDa. Hyperimmune gnotobiotic pig serum raised against purified axial filaments of strain P18A (serotype 4) cross-reacted with all other serotypes and with the non-pathogens, and convalescent serum taken from a pig with persistent swine dysentery also showed a strong response to the axial filament polypeptides. Hyperimmune gnotobiotic pig serum raised against axial filaments failed to agglutinate viable organisms and did not inhibit growth in vitro. Hence, the axial filaments of T. hyodysenteriae have been identified as major immunodominant antigens, although the role that antibodies to these antigens play in protection has yet to be established.

Sonicated preparations of Treponema hyodysenteriae and Treponema innocens were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis. Treponemal proteins were electrophoresed on a 10% polyacrylamide slab gel in a discontinuous Tris-glycine system and either stained with Coomassie blue dye or transferred electrophoretically at 20 mA for 16 h and 30 mA for 3 h to nitrocellulose paper. Staining of the gels revealed at least 42 distinct T. hyodysenteriae and T. innocens proteins, with molecular sizes ranging from greater than 100 to 14 kilodaltons (kDa). Each species contained 12 to 16 major protein bands; five of the proteins were common to both species. Fourteen major antigens were identified in T. hyodysenteriae isolate B204 by using serum specimens from pigs in the acute stage of swine dysentery. Twelve additional antigens were detected in isolate B204 when convalescent-phase serum specimens were reacted to the blot. A wide band at 16 kDa was identified with convalescent-phase serum specimens in T. hyodysenteriae but not in T. innocens. This 16-kDa antigen was also identified in T. hyodysenteriae with colonic secretions from convalescent pigs.

Borrelia burgdorferi strains (six isolates from North America and 28 isolates from Europe) were analyzed by physicochemical and immunological methods. By SDS-PAGE, all Borrelia burgdorferi strains tested had two major proteins with constant molecular weights of 60 and 41 kDa and one, two, or three variable low molecular weight proteins (OspA = 30-32 kDa, OspB = 34-36 kDa, pC = 21-22 kDa). All combinations--except OspB alone or OspB/pC--were observed. Borrelia burgdorferi strains were different from relapsing fever borreliae by strong reactivity with OspA- and/or pC-specific polyclonal antibodies, whereas relapsing fever borreliae were only weakly reactive. Among 25 Borrelia burgdorferi isolates, seven different serotypes of Borrelia burgdorferi were defined according to their reactivity in the Western blot with three monoclonal OspA-specific antibodies (H5332, H3TS, and LA5), four OspA- or OspB-specific polyclonal antibodies, and 12 polyclonal antibodies against whole borreliae. Antigenic differences between European CSF and skin isolates were observed, all skin isolates (n = 11) belonging to serotype 2 in contrast to only two out of seven CSF isolates. CSF isolates were antigenically heterogenous (serotypes 1, 2, 3, 4, and 5). Serotypes 6 and 7 were represented by two tick isolates, and the other European tick isolates are not yet fully characterized. Antigenic differences between European and North American strains may play a role in differences in the clinical picture of Lyme borreliosis.