Separation of Treponema pallidum from Tissue Substances by Continuous-Flow Zonal Centrifugation

The zonal ultracentrifuge was used for separation of Treponema pallidum from large volumes of rabbit testicular syphiloma extracts by continuous-flow centrifugation in a cesium chloride density gradient. The gradient was linear with radius from a density of 1.05 to 1.36 g/ml. Operating speeds were 15,000 rev/min for the continuous-flow phase and 25,000 rev/min for a 30-min banding period. A total of 9 x 10(9) (24.3%) treponemes were recovered from the original extract. Of the treponemes recovered, 88% formed a band at a density of 1.170 to 1.211 g/ml. Within the limits of present methods of assay, these fractions were relatively free from testicular particulates and protein when compared with treponemes recovered after differential centrifugation. Observations of the isolated fractions by dark-field and electron microscopy indicated a lack of gross morphological damage to T. pallidum. Their antigenic characteristics were also retained, as evidenced by their ability to react with syphilitic sera in the indirect fluorescent-antibody procedure.     

Phylogenetic analysis of the spirochetes.

The 16S rRNA sequences were determined for species of Spirochaeta, Treponema, Borrelia, Leptospira, Leptonema, and Serpula, using a modified Sanger method of direct RNA sequencing. Analysis of aligned 16S rRNA sequences indicated that the spirochetes form a coherent taxon composed of six major clusters or groups. The first group, termed the treponemes, was divided into two subgroups. The first treponeme subgroup consisted of Treponema pallidum, Treponema phagedenis, Treponema denticola, a thermophilic spirochete strain, and two species of Spirochaeta, Spirochaeta zuelzerae and Spirochaeta stenostrepta, with an average interspecies similarity of 89.9%. The second treponeme subgroup contained Treponema bryantii, Treponema pectinovorum, Treponema saccharophilum, Treponema succinifaciens, and rumen strain CA, with an average interspecies similarity of 86.2%. The average interspecies similarity between the two treponeme subgroups was 84.2%. The division of the treponemes into two subgroups was verified by single-base signature analysis. The second spirochete group contained Spirochaeta aurantia, Spirochaeta halophila, Spirochaeta bajacaliforniensis, Spirochaeta litoralis, and Spirochaeta isovalerica, with an average similarity of 87.4%. The Spirochaeta group was related to the treponeme group, with an average similarity of 81.9%. The third spirochete group contained borrelias, including Borrelia burgdorferi, Borrelia anserina, Borrelia hermsii, and a rabbit tick strain. The borrelias formed a tight phylogenetic cluster, with average similarity of 97%. THe borrelia group shared a common branch with the Spirochaeta group and was closer to this group than to the treponemes. A single spirochete strain isolated fromt the shew constituted the fourth group. The fifth group was composed of strains of Serpula (Treponema) hyodysenteriae and Serpula (Treponema) innocens. The two species of this group were closely related, with a similarity of greater than 99%. Leptonema illini, Leptospira biflexa, and Leptospira interrogans formed the sixth and most deeply branching group. The average similarity within this group was 83.2%. This study represents the first demonstration that pathogenic and saprophytic Leptospira species are phylogenetically related. The division of the spirochetes into six major phylogenetic clusters was defined also by sequence signature elements. These signature analyses supported the conclusion that the spirochetes represent a monophylectic bacterial phylum.     

Characterization of monoclonal antibodies to Treponema pallidum

Thirteen hybrid cell lines which produce mouse monoclonal antibodies to Treponema pallidum, the causative agent of syphilis, have been established. All of the monoclonal antibodies react with T. pallidum, Nichols strain, in ELISA and in immunofluorescence assays, but do not react with normal rabbit testicular tissue in the ELISA. Two of these antibodies were demonstrated to react with the nonpathogenic treponemes T. phagedenis, biotype Reiter, T. refringens (Noguchi strain), T. vincentii, and T. denticola (strains 11 and W), as well as with Borrelia recurrentis, Leptospira interrogans, serogroup Canicola, and the swine pathogen T. hyodysenteriae. The remaining 11 antibodies react with four recently isolated strains of T. pallidum, but with none of the related nonpathogens nor with Borrelia or Leptospira. Thus, our results to date indicate that these monoclonal antibodies may identify antigenic determinants that are specific either for T. pallidum alone or for those treponemes which are pathogenic for humans. The molecular specificities of six of the 13 antibodies were determined by Western blotting. We anticipate potential usefulness of these antibodies in the investigation of the antigenic structure of T. pallidum, the taxonomic study of the pathogenic and nonpathogenic treponemes, and in the diagnosis of syphilis.     

DNA characterization of Lyme disease spirochetes.

Lyme disease spirochetes (LDS) have phenotypic characteristics of both treponemes and borreliae. To ascertain whether one or more species of LDS exist, as well as their taxonomic status, we determined the DNA base (G + C) content for three strains of LDS, the DNA relatedness of ten strains isolated in the United States or Europe, and the DNA relatedness of LDS to other spirochetes. The G + C content of the three LDS strains was 28.1-29.0 mol%, most similar to those of Borellia hermsii (30.6 mol %) and Treponema hyodysenteriae (25.6 mol %) among the other spirochetes tested. DNA hybridization studies of nine LDS strains to a reference strain isolated from human blood revealed divergence (unpaired bases) within related nucleotide sequences of only 0.0-1.0 percent, indicating the strains were one species. Similarly, relatedness values of seven strains to the reference strain were high: 58-98 percent (mean, 71 percent) in 50 degrees C reactions and 50-93 percent (mean, 69 percent) in 65 degrees C reactions. Labeled DNA from B. hermsii was 30-40 percent related to three Lyme disease spirochete strains in 50 degrees C reactions and 8-10 percent related in 65 degrees C reactions. In contrast, DNA from the reference LDS strain showed relatedness of only 1 percent to DNAs of two leptospires and only 16 percent to DNA from T. hyodysenteriae. We conclude that LDS are a single species, genetically unlike treponemes or leptospires, which belong in the genus Borrelia.     

Scanning electron microscopy of Treponema pallidum (Nichols strain) attached to cultured mammalian cells.

This paper describes the attachment of Treponema pallidum (Nichols strain) to cultured mammalian cells as a visualized by scanning electron microscopy. Treponemes were incubated for 3 hr with cultured cells derived from normal rabbit testes or human skin epithelium, then fixed, processed with critical-point drying, and examined with a Cambridge Mark 2A scanning electron microscope. Large numbers of treponemes became attached to the cultured cells without altering the morphological integrity of the cultured cells. Attachment appeared to involve a very close physical proximity of treponemes to the cultured cells; at the site of attachment, no changes such as swelling or indentation of the cultured cell surface were observed. The addition of ruthenium red to the fixatives produced a treponemal-associated surface precipitate. This material, which is probably mucopolysaccharide and/or phospholipid, may be important in protecting the organisms against host defense mechanisms; in addition, it may be involved in the serological unresponsiveness of freshly prepared suspensions of T. pallidum.     

Treponema pallidum rare outer membrane proteins: analysis of mobility by freeze-fracture electron microscopy.

Freeze-fracture and deep-etch electron microscopy were used to investigate the molecular architecture of the Treponema pallidum outer membrane (OM). Freeze-fracture electron microscopy of treponemes freshly harvested from rabbit testes revealed that the intramembranous particles (IMPs) in both the concave and convex OM leaflets were distributed into alternating areas of relatively high and low particle density; in many OM fractures, IMPs formed rows that ran either parallel to or obliquely across the fracture faces. Statistical analysis (runs test) confirmed that the IMPs were nonrandomly distributed in both OM leaflets. Examination of deep-etched specimens revealed that the particles observed in freeze-fractured OMs also were surface exposed. Combined analysis of deep-etched and cross-fractured treponemes revealed that the OM particles were located in regions of the OM away from the endoflagella and closely apposed to the cytoplasmic membrane-peptidoglycan complex. When treponemes were incubated for extended periods with heat-inactivated immune rabbit syphilitic serum, no alteration in the distribution of OM IMPs was detected. In further experiments, approximately 1:1 mixtures of T. pallidum and Escherichia coli or separate suspensions of the nonpathogenic Treponema phagedenis biotype Reiter were fixed at 34 degrees C or after cooling to 0 degree C (to induce lateral phase separations that would aggregate IMPs). Only particles in the T. pallidum OM failed to aggregate in cells fixed at the lower temperature. The combined data suggest that the mobility of T. pallidum rare OM proteins is limited, perhaps as a result of interactions between their periplasmic domains and components of the peptidoglycan-cytoplasmic membrane complex.     

A murine IgM monoclonal antibody binds an antigenic determinant in outer surface protein A, an immunodominant basic protein of the Lyme disease spirochete

A hybridoma cell line formed by the fusion of the P3x63-Ag8.653 myeloma cell line with splenocytes from BALB/c mice immunized with Borrelia burgdorferi produced an IgM monoclonal antibody (mAb-11G1) with kappa-light chains which detected an antigenic determinant in a major spirochetal protein of m.w. approximately 31,000, also known as outer surface protein A (OSP-A). Apparent saturation was reached in approximately 35 min with 34 ng of mAb-11G1 binding to 5 X 10(7) spirochetes giving an estimated 4.8 X 10(2) IgM molecules per spirochete and thus a minimum of 480 binding sites per organism. Enzymatic digestion studies suggest that the antigenic determinant to mAb-11G1 is contained within the peptide chain of OSP-A as binding could be eliminated by treatment of the spirochetes with proteinase K, Pronase and pepsin (100 to 200 micrograms/ml of enzyme) but not by trypsin or bromelain treatment. Periodate oxidation as well as mixed and endoglycosidase treatment of the spirochetes did not alter the binding of mAb-11G1. Two-dimensional gel electrophoresis of whole spirochetal cell lysates disclosed that OSP-A is a heterogeneously charged basic protein with an apparent isoelectric point range from 8.5 to 9.0. Amino acid analysis of OSP-A showed a 10% lysine component which could provide the basic nature to the protein. OSP-A with the intact antigenic determinant for mAb-11G1 can be found in the urine of hamsters experimentally infected with B. burgdorferi.     

Complement activation limits the rate of in vitro treponemicidal activity and correlates with antibody-mediated aggregation of Treponema pallidum rare outer membrane protein

A modification of the in vitro immobilization assay together with freeze-fracture analysis was used to determine the factors responsible for the prolonged time required in vitro to achieve killing of Treponema pallidum subsp. pallidum. The modified immobilization assay permitted separate determination of the time required for binding of antibody to the surface of T. pallidum and for C activation. Treponemes were preincubated in heat-inactivated immune rabbit serum (IRS) followed by washing the organisms in 2.5% BSA/PBS to remove unbound IRS antibody before the addition of C. The results showed that a comparable degree of C-dependent killing occurred when treponemes were preincubated in heat-inactivated IRS for either 30 min or 16 h, indicating that treponemicidal antibody rapidly binds to the surface of T. pallidum. Preincubation of treponemes for 17 h in heat-inactivated IRS followed by a 1-h incubation in C resulted in the loss of 80% treponemal motility, indicating that C activation results in rapid killing of T. pallidum. Treponemes preincubated in IRS for 1 h, then incubated for 8 h and 16 h in heat-inactivated normal serum also lost a significant level of motility after the addition of C; in contrast, motility was unaffected after 30 min and 4 h of incubation in heat-inactivated normal serum under similar conditions. These results demonstrate that, whereas antibody binding to and C-mediated killing of treponemes can proceed rapidly, the prolonged time to C activation limits the rate at which treponemicidal activity occurs in vitro. In addition, treponemicidal activity using the modified immobilization assay could not be demonstrated with antiserum against T. pallidum endoflagella, antiserum against proteins solubilized from T. pallidum using the detergent Triton X-114, and a mAb to the T. pallidum r190-kDa "4D" protein, suggesting that these molecules are not accessible to surface binding antibody. Freeze-fracture analysis, recently used in our laboratory to demonstrate that the outer membrane of T. pallidum has rare constituent protein, was utilized to demonstrate outer membrane target Ag of IRS antibody. T. pallidum rare outer membrane protein (TROMP) molecules were shown in freeze-fracture electron micrographs to be consistently aggregated following a 16-h incubation of treponemes in IRS. In contrast, no aggregation of TROMP was present in treponemes incubated in normal rabbit serum for 16 h or in treponemes incubated in IRS for 2 h. These findings suggest that the rate of C activation leading to in vitro treponemicidal activity is limited by the time required for aggregation of antibody-bound TROMP molecules.     

Antioxidants attenuate endotoxin-induced microvascular leakage of macromolecules in vivo.

The purpose of this study was to examine whether antioxidants attenuate endotoxin-induced microvascular hyper-permeability for macromolecules in the hamster cheek pouch. Twenty-two adult male Syrian hamsters were anesthetized, and a removable plastic chamber was placed in the cheek pouch to observe and collect suffusate from the microvasculature. Fluorescent-labeled dextran (FITC-D; mol wt 150,000) was injected intravenously, and changes in the number of microvascular leaky sites and microvascular clearance of FITC-D were measured in five groups: saline control (group 1, n = 4), endotoxin (0.1 mg/ml) suffusion for 120 min (group 2, n = 6), endotoxin plus dimethyl sulfoxide (1.0 g/kg iv; group 3, n = 4), endotoxin plus allopurinol (30 mg/kg ip; group 4, n = 4), and endotoxin plus dimethyl sulfoxide and allopurinol (group 5, n = 4). The number of leaky sites and the FITC-D clearance were significantly higher in group 2 [45 +/- 18 (SD) sites/cm2 and 20 +/- 6 X 10(-6) ml/min, respectively; P less than 0.01] than in group 1 (7 +/- 6 sites/cm2 and 7 +/- 5 X 10(-6) ml/min), group 3 (9 +/- 5 sites/cm2 and 8 +/- 2 X 10(-6) ml/min), group 4 (11 +/- 7 sites/cm2 and 9 +/- 4 X 10(-6) ml/min), and group 5 (11 +/- 6 sites/cm2 and 7 +/- 1 x 10(-6) ml/min). The leaky sites appeared predominantly in postcapillary venules. There was a positive and significant correlation between the number of leaky sites and FITC-D clearance.(ABSTRACT TRUNCATED AT 250 WORDS)     

Treponemicidal activity of anti-treponemal lymphotoxins and their relation to circulating immune complexes and autolymphocytotoxins

Abstract It was previously reported that spleen cells of rabbits infected with Treponema pallidum produced anti-treponemal lymphotoxins (ATL). This ability was distinctly disturbed when circulating immune complexes (CIC) and autolymphocytotoxins (AL) were present in the sera of cell donors. ATL liberated from cells of donors without CIC and AL displayed a marked ability to immobilize treponemes. The percentage of immobilized treponemes varied according to the type of cells used for ATL liberation and their density. The most active was ATL from T cells (density 4 × 10⁸ ml⁻¹) and the weakest was the one from B lymphocytes. In the presence of CIC in sera of cell donors the weakest ATL was from macrophages and in the presence of AL from T lymphocytes. When both factors (CIC and AL) were present ATL from T lymphocytes did not immobilize treponemes. This seems to suggest that the impairment of the cells' ability to produce ATL may facilitate the survival of treponemes in the host despite the presence of immunologically competent cells.     

Fatty acids of Treponema pallidum and Borrelia burgdorferi lipoproteins.

A fundamental ultrastructural feature shared by the spirochetal pathogens Treponema pallidum subsp. pallidum (T. pallidum) and Borrelia burgdorferi, the etiological agents of venereal syphilis and Lyme disease, respectively, is that their most abundant membrane proteins contain covalently attached fatty acids. In this study, we identified the fatty acids covalently bound to lipoproteins of B. burgdorferi and T. pallidum and examined potential acyl donors to these molecules. Palmitate was the predominant fatty acid of both B. burgdorferi and T. pallidum lipoproteins. T. pallidum lipoproteins also contained substantial amounts of stearate, a fatty acid not typically prevalent in prokaryotic lipoproteins. In both spirochetes, the fatty acids of cellular lipids differed from those of their respective lipoproteins. To characterize phospholipids in these organisms, spirochetes were metabolically labeled with [3H]palmitate or [3H]oleate; B. burgdorferi contained only phosphatidylglycerol and phosphatidylcholine, while T. pallidum contained phosphatidylglycerol, phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, and cardiolipin. Although palmitate predominated in the lipoproteins, there were no apparent differences in the incorporation of these two fatty acids into phospholipids (putative acyl donors). Phospholipase A1 and A2 digestion of phosphatidylcholine from B. burgdorferi and T. pallidum labeled with either [3H]palmitate or [3H]oleate also revealed that neither fatty acid was incorporated preferentially into the 1 and 2 positions (potential acyl donor sites) of the glycerol backbone. The combined findings suggest that fatty acid utilization during lipoprotein synthesis is determined largely by the fatty acid specificities of the lipoprotein acyl transferases. These findings also provide the basis for ongoing efforts to elucidate the relationship between lipoprotein acylation and the physiological functions and inflammatory activities of these molecules.     

The effect of reducing and other agents on the motility of Treponema pallidum in an acellular culture medium.

The maintenance of Treponema pallidum motility was investigated in an acellular medium based on T. pallidum immobilization test medium. The acellular medium contained cysteine, glutathione, thioglycollate and dithiothreitol as reducing agents and had a redox potential of -275 +/- 25 mV at pH 7.3. In an atomosphere containing 3% O2, motile treponemes survived four times longer when calf serum and bovine serum albumin were added to the medium. The selective omission of glutathione and, particularly, thioglycollate prolonged the survival of motile treponemes almost fivefold. In addition, stored medium, in which thioglycollate had become inactive, sustained motile treponemes for longer than did freshly prepared medium. Thus, thioglycollate is toxic for the organisms. It may be omitted from the medium because low redox potentials can be achieved without it.     

Removal of bacteria from water by adhesion to cross-linked poly(vinylpyridinium halide).

Cross-linked poly(vinylpyridinium halide) was found to have a novel and remarkable ability to remove bacteria from water. For example, when 10 g (wet weight) of cross-linked poly(N-benzyl-4-vinylpyridinium bromide) was contacted with 20 ml of suspensions of Escherichia coli (9.7 X 10(4) to 9.7 X 10(7)/ml), Salmonella typhimurium (8.0 X 10(6) to 1.1 X 10(7)/ml), Streptococcus faecalis (5.0 X 10(7)/ml), Staphylococcus aureus (8.1 X 10(7)/ml), and Pseudomonas aeruginosa (3.2 X 10(5)/ml) under stirring in sterilized physiological saline at 37 degrees C, 99% of the viable cells of these bacteria were removed in 2 to 6 h. When suspensions of these bacteria (10(5) to 10(8) cells per ml) were passed through a column (20 mm by 100 cm) of cross-linked poly(N-benzyl-4-vinylpyridinium bromide) at 37 degrees C with a flow rate of 0.8 to 1.4 bed volumes per h, 97 to 100% of the viable cells were eliminated from the suspensions during the treatment. Mechanistic studies demonstrated that cross-linked poly(vinylpyridinium halide) irreversibly captured these bacteria alive during the treatment. That is, total organic carbon was removed during the treatment, and the bacteria which adhered to the resin proliferated on the bacterial medium. The adhesion capacity was estimated to be 10(10) cells per g (dry weight). Total organic carbon was also removed even when the bacteria were killed by heat treatment before the column studies.     

The outer membrane of Treponema pallidum: biological significance and biochemical properties

Rabbits infected intravenously with Treponema pallidum were not markedly febrile, and the pyrogenicity of treponeme preparations administered intravenously to rabbits was negligible. The antibiotic polymyxin B did not induce any ultrastructural changes on the treponemal surface and was not lethal (immobilizing) for T. pallidum, which was, however, highly susceptible to detergents such as SDS. Extraction of treponemes with Triton X-100 removed the outer membrane (despite the presence of Mg2+) as shown by electron microscopy, and solubilized a limited number of proteins detectable by SDS-PAGE, including a dominant antigen (47 kDal) demonstrated by immunoblotting. None of the proteins were heat-modifiable. Periodic acid-silver staining of polyacrylamide gels for carbohydrate together with protease K digestion did not demonstrate major carbohydrate components in whole treponemes, or in the Triton-soluble fraction. Surface iodination of intact treponemes revealed very little surface exposure of treponemal proteins, although a protein which co-migrated with host albumin was labelled and appeared to be associated with the treponemal surface. Many treponemal proteins were, however, labelled when iodination was done in the presence of Triton. These observations, indicate that the outer membrane of T. pallidum differs significantly from those of many Gram-negative pathogens.     

Characterization of the serum requirement for macrophage-mediated killing of Treponema pallidum ssp. pallidum: Relationship to the development of opsonizing antibodies

Abstract Serum pools were collected from rabbits bled at various times after intra-testicular infection with Treponema pallidum ssp. pallidum . These were tested for their ability to opsonize T. pallidum and promote killing of the organisms by macrophages. Compared to normal sera, significant opsonization was first seen on day 10 of infection as measured by both ingestion ( P < 0.001) and macrophage-mediated killing ( P = 0.006); significant levels of functional antibodies persisted through 300 days of infection. Although opsonic activity peaked early in infection, antibodies that promoted optimal macrophage-mediated killing developed much later, suggesting that these two functions may represent activities of antibodies with differing specificities or affinities. The initial development of antibodies that augment both phagocytosis and killing corresponds with the in vivo clearance of treponemes from the primary site of infection. These observations support the hypothesis that macrophages are the major effector mechanism for elimination of T. pallidum during early syphilis infection.     

Analysis of outer membrane ultrastructure of pathogenic Treponema and Borrelia species by freeze-fracture electron microscopy.

We analyzed the outer membrane (OM) ultrastructure of four pathogenic members of the family Spirochaetaceae by freeze fracture. The OM of Treponema pallidum subsp. pertenue contained a low intramembranous particle concentration, indicating that it contains few OM transmembrane proteins. The concave OM fracture faces of Treponema hyodysenteriae and Borrelia burgdorferi contained dense populations of particles, typical of gram-negative organisms. A relatively low concentration of particles which were evenly divided between a small and a large species was present in the concave OM fracture face of Borrelia hermsii; the convex OM fracture face contained only small particles. As for gram-negative bacteria, the convex OM fracture face particle concentrations of these pathogens were low. These spirochetes cleaved preferentially within the OM, in contrast to typical gram-negative bacteria, which tend to fracture within the inner membrane. The OM ultrastructure of T. pallidum subsp. pertenue provides an explanation for the lack of antigenicity of the treponemal surface and may reflect a mechanism by which this pathogen evades the host immune response.     

Use of the polymerase chain reaction to detect DNA sequences specific to pathogenic treponemes in cerebrospinal fluid

The polymerase chain reaction was used to detect Treponema pallidum in specimens of cerebrospinal fluid (CSF), as a means of diagnosing syphilis. Segments of the TmpA and 4D genes were amplified to provide an estimated threshold sensitivity of approximately 65 organisms in 0.5 ml. A spectrum of pathogens known to cause meningitis, and several non-pathogenic treponemes were unreactive. Treponema pertenue, and only one of 30 control specimens of CSF were positive. In contrast, 10 of 19 CSFs from patients being evaluated for latent or tertiary syphilis were positive, as were 7 of 28 specimens from HIV-positive patients.     

Demonstration of rare protein in the outer membrane of Treponema pallidum subsp. pallidum by freeze-fracture analysis.

The surface of Treponema pallidum subsp. pallidum (T. pallidum), the etiologic agent of syphilis, appears antigenically inert and lacks detectable protein, as judged by immunocytochemical and biochemical techniques commonly used to identify the outer membrane (OM) constituents of gram-negative bacteria. We examined T. pallidum by freeze-fracture electron microscopy to visualize the architecture of its OM. Treponema phagedenis biotype Reiter (T. phagedenis Reiter), a nonpathogenic host-associated treponeme, and Spirochaeta aurantia, a free-living spirochete, were studied similarly. Few intramembranous particles interrupted the smooth convex and concave fracture faces of the OM of T. pallidum, demonstrating that the OM of this organism is an unusual, nearly naked lipid bilayer. In contrast, the concave fracture face of the OM of S. aurantia was densely covered with particles, indicating the presence of abundant integral membrane proteins, a feature shared by typical gram-negative organisms. The concentration of particles in the OM concave fracture face of T. phagedenis Reiter was intermediate between those of T. pallidum and S. aurantia. Similar to typical gram-negative bacteria, the OM convex fracture faces of the three spirochetes contained relatively few particles. The unique molecular architecture of the OM of T. pallidum can explain the puzzling in vitro properties of the surface of the organism and may reflect a specific adaptation by which treponemes evade the host immune response.     

Glycerol-3-phosphate acquisition in spirochetes: distribution and biological activity of glycerophosphodiester phosphodiesterase (GlpQ) among Borrelia species

Relapsing-fever spirochetes achieve high cell densities (>10(8)/ml) in their host's blood, while Lyme disease spirochetes do not (<10(5)/ml). This striking contrast in pathogenicity of these two groups of bacteria suggests a fundamental difference in their ability to either exploit or survive in blood. Borrelia hermsii, a tick-borne relapsing-fever spirochete, contains orthologs to glpQ and glpT, genes that encode glycerophosphodiester phosphodiesterase (GlpQ) and glycerol-3-phosphate transporter (GlpT), respectively. In other bacteria, GlpQ hydrolyzes deacylated phospholipids to glycerol-3-phosphate (G3P) while GlpT transports G3P into the cytoplasm. Enzyme assays on 17 isolates of borreliae demonstrated GlpQ activity in relapsing-fever spirochetes but not in Lyme disease spirochetes. Southern blots demonstrated glpQ and glpT in all relapsing-fever spirochetes but not in the Lyme disease group. A Lyme disease spirochete, Borrelia burgdorferi, that was transformed with a shuttle vector containing glpTQ from B. hermsii produced active enzyme, which demonstrated the association of glpQ with the hydrolysis of phospholipids. Sequence analysis of B. hermsii identified glpF, glpK, and glpA, which encode the glycerol facilitator, glycerol kinase, and glycerol-3-phosphate dehydrogenase, respectively, all of which are present in B. burgdorferi. All spirochetes examined had gpsA, which encodes the enzyme that reduces dihydroxyacetone phosphate (DHAP) to G3P. Consequently, three pathways for the acquisition of G3P exist among borreliae: (i) hydrolysis of deacylated phospholipids, (ii) reduction of DHAP, and (iii) uptake and phosphorylation of glycerol. The unique ability of relapsing-fever spirochetes to hydrolyze phospholipids may contribute to their higher cell densities in blood than those of Lyme disease spirochetes.     

Membrane cartridges for endotoxin removal from interferon preparations

The suitability of membrane cartridges for the removal of endotoxin from both distilled water and interferon preparations was examined. The endotoxin concentrations were reduced to 4.0 and 7.3 EU/ml, respectively, when about 4000 ml of distilled water with 20 and 28 EU/ml were passed through the deoxycholate and chitosan immobilized membrane cartridges. When 200 ml of interferon preparation with endotoxin concentration more than 80 EU/ml and pH 3.9 were applied to a deoxycholate immobilized membrane cartridge at a flow-rate of 9 ml/min, the endotoxin concentration was reduced to less than 10 EU/ml. However, if an interferon preparation of 450 ml, with more than 80 EU/ml of endotoxin and pH 3.9 was applied to the chitosan immobilized membrane cartridge at a flow-rate of 18 ml/min, the endotoxin concentration was reduced to less than 10 EU/ml.