Plasmid analysis of Borrelia burgdorferi, the Lyme disease agent.

A simple procedure for extraction of plasmid-enriched DNA from borreliae was used in a plasmid analysis of 13 strains of the Lyme disease agent, Borrelia burgdorferi. The extracted DNA was subjected to low-percentage agarose gel electrophoresis and examined either directly by ethidium bromide staining or after hybridization of the plasmids in situ with a DNA probe for the gene encoding the major outer membrane protein OspA. Each isolate had four to seven discernible plasmids of various sizes. Only 2 of the 13 strains had the same plasmid profile. The ospA gene probe hybridized to large plasmids to strains from both North America and Europe. A strain which had been passaged many times was found to have lost two of the six plasmids originally present. These findings indicate the potential usefulness of plasmid analysis as a strain-typing procedure and for identifying possible plasmid-conferred virulence factors.     

Lyme disease assay which detects killed Borrelia burgdorferi.

We developed an in vitro assay showing that Borrelia burgdorferi organisms were killed by serum from patients with Lyme disease. Twenty of 20 Lyme disease serum samples caused B. burgdorferi killing in a range of 36 to 99% compared with the mean number of viable spirochetes when sera from 10 healthy individuals were used. The percentage of killing of B. burgdorferi increased with convalescent serum from patients with early Lyme disease. The borreliacidal activity was detectable in some sera diluted 640-fold and was abrogated after treatment with anti-human immunoglobulin G. In contrast, pooled or individual normal human serum did not cause a decrease in the number of viable B. burgdorferi. Borreliacidal activity was also not detected in sera from patients with relapsing fever, rocky mountain spotted fever, syphilis, mononucleosis, rheumatoid factor, or DNA antibodies. Our results show that borreliacidal activity can be used as a specific serodiagnostic test for detecting Lyme disease.     

Absence of lipopolysaccharide in the Lyme disease spirochete, Borrelia burgdorferi

We were unable to demonstrate the presence of the classic enterobacterium-type lipopolysaccharide in the cells of the Lyme disease spirochete, Borrelia burgdorferi B31. This finding was primarily based on chemical analysis and the absence of free lipid A upon mild acid hydrolysis of the appropriate cell extracts. These results do not preclude the possible existence of an unusual lipopolysaccharide-like compound(s) in B. burgdorferi.     

Amplification of Borrelia burgdorferi DNA in skin biopsies from patients with Lyme disease.

To determine whether the polymerase chain reaction could contribute to a better diagnosis of Lyme disease, skin biopsy samples from patients suffering from erythema chronicum migrans or acrodermatitis chronica atrophicans were tested for the presence of Borrelia burgdorferi by a polymerase chain reaction assay, which was specific for European strains. The spirochete could not be detected microscopically in any of the 15 biopsy samples obtained from nine patients. However, B. burgdorferi could be isolated from seven of eight of these samples, which indicated the presence of spirochetes. Using a nested polymerase chain reaction, we were able to detect B. burgdorferi-specific sequences in 12 of the 15 biopsy samples. Biopsy samples from three of four patients with erythema chronicum migrans and four of five patients with acrodermatitis chronica atrophicans were found to be positive for B. burgdorferi. The spirochete could be isolated from the biopsy sample, from a patient with erythema chronicum migrans who tested negative, which suggests a false-negative polymerase chain reaction result probably on account of the low number of spirochetes present in the lesion. The positive polymerase chain reaction for lesions from patients with acrodermatis chronica atrophicans supports the concept that B. burgdorferi can persist in the skin over a long period of time. From these results, it was concluded that the polymerase chain reaction is a valuable technique for the diagnosis of Lyme disease.     

Hemagglutination and proteoglycan binding by the Lyme disease spirochete, Borrelia burgdorferi.

The ability of the Lyme disease spirochete to attach to host components may contribute to its ability to infect diverse tissues. We present evidence that the Lyme disease spirochete expresses a lectin activity that promotes agglutination of erythrocytes and bacterial attachment to glycosaminoglycans. Among a diverse collection of 21 strains of Lyme disease spirochete, hemagglutinating activity was easily detected in all but 3 strains, and these three strains were noninfectious. The ability to agglutinate erythrocytes was associated with the ability of the spirochete to bind to the sulfated polysaccharide dextran sulfate and to mammalian cells. Soluble dextran sulfate was a potent inhibitor of both hemagglutination and attachment to mammalian cells, while dextran had no effect on either activity, suggesting that dextran sulfate may inhibit attachment by mimicking host cell glycosaminoglycans. Consistent with this, the spirochete bound to immobilized heparin, and soluble heparin inhibited bacterial adhesion to mammalian cells. The bacterium did not bind efficiently to Vero cells treated with heparinase or heparitinase or to mutant CHO cell lines that are deficient in proteoglycan synthesis. Sulfation of glycosaminoglycans was critical for efficient bacterial recognition, as Vero cells treated with an inhibitor of sulfation, or a mutant CHO cell line that produces undersulfated heparan sulfate, did not mediate maximal spirochetal binding. Binding of the spirochete to extracellular matrix also appeared to be dependent upon this attachment pathway. These findings suggest that a glycosaminoglycan-binding activity which can be detected by hemagglutination contributes to the attachment of the Lyme disease spirochete to host cells and matrix.     

Characterization of the first tick isolate of Borrelia burgdorferi from Italy

We report on the first isolation of a spirochetal organism from Ixodes ricinus ticks of the Trieste area (Northern Italy) which was identified as Borrelia burgdorferi by its reactivity with specific monoclonal antibodies directed against the OSPA and flagella proteins.     

Characterization of outer membranes isolated from Borrelia burgdorferi, the Lyme disease spirochete.

The lack of methods for isolating Borrelia burgdorferi outer membranes (OMs) has hindered efforts to characterize borrelial surface-exposed proteins. Here we isolated OMs by immersion of motile spirochetes in hypertonic sucrose followed by isopycnic ultracentrifugation of the plasmolyzed cells. The unilamellar vesicles thus obtained were shown to be OMs by the following criteria: (i) they contained OspA and OspB; (ii) they did not contain flagellin, NADH oxidase activity, or the 60-kDa heat shock protein; and (iii) their morphology by freeze-fracture electron microscopy was identical to that of OMs of intact organisms. Consistent with previous studies which employed immunoelectron microscopy and detergent-based solubilization of B. burgdorferi OMs, only small proportions of the total cellular content of OspA or OspB were OM associated. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) fluorography of OMs from spirochetes metabolically radiolabeled with [3H]palmitate or 35S-amino acids demonstrated that the OMs contained both nonlipidated and lipidated proteins. This fractionation procedure was also used to isolate OMs from virulent and avirulent isolates of the well-characterized B. burgdorferi N40 strain. SDS-PAGE fluorography revealed that OMs from the two isolates differed with respect to both nonlipoprotein and lipoprotein constituents. When whole cells, protoplasmic cylinders, and OMs were immunoblotted against sera from mice persistently infected with B. burgdorferi N40, the majority of antibody reactivity was directed against intracellular proteins. The availability of isolated OMs should facilitate efforts to elucidate the complex relationship(s) between B. burgdorferi membrane composition and Lyme disease pathogenesis.     

Temperature-related differential expression of antigens in the Lyme disease spirochete, Borrelia burgdorferi.

Previous studies have demonstrated that Borrelia burgdorferi in the midguts of infected ticks shows increased expression of the antigenic outer surface protein OspC after the ticks have ingested a blood meal. This differential expression is at least partly due to a change in temperature, as an increase in OspC levels is also observed when cultures are shifted from 23 to 35 degrees C. Immunoblotting of bacterial lysates with sera from infected mice indicated that the levels of several additional antigens were also increased in bacterial cultures shifted to 35 degrees C; we have identified one antigen as OspE. We have also observed differential expression of OspF, which has been proposed to be coexpressed in an operon with the gene encoding OspE.     

Distribution and molecular analysis of Lyme disease spirochetes, Borrelia burgdorferi, isolated from ticks throughout California.

Previous studies describing the occurrence and molecular characteristics of Lyme disease spirochetes, Borrelia burgdorferi, from California have been restricted primarily to isolates obtained from the north coastal region of this large and ecologically diverse state. Our objective was to look for and examine B. burdorferi organisms isolated from Ixodes pacificus ticks collected from numerous regions spanning most parts of California where this tick is found. Thirty-one isolates of B. burgdorferi were examined from individual or pooled I. pacificus ticks collected from 25 counties throughout the state. One isolate was obtained from ticks collected at Wawona Campground in Yosemite National Park, documenting the occurrence of the Lyme disease spirochete in an area of intensive human recreational use. One isolate from an Ixodes neotomae tick from an additional county was also examined. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblot analysis, agarose gel electrophoresis, Southern blot analysis, and the polymerase chain reaction were used to examine the molecular and genetic determinants of these uncloned, low-passage-number isolates. All of the isolates were identified as B. burgdorferi by their protein profiles and reactivities with monoclonal and polyclonal antibodies, and all the isolates were typed by the polymerase chain reaction as North American-type spirochetes (B. burgdorferi sensu stricto). Although products of the ospAB locus were identified in protein analyses in all of the isolates, several isolates contained deleted forms of this locus that would result in the expression of chimeric OspA-OspB proteins. The analysis of OspC demonstrated that this protein was widely conserved among the isolates but was also quite variable in its molecular mass and the amount of it that was expressed.     

Demonstration of a B-lymphocyte mitogen produced by the Lyme disease pathogen, Borrelia burgdorferi.

Lyme disease refers to the multisymptomatic illness in humans which results from infection with the tick-borne spirochete Borrelia burgdorferi. The white-footed mouse is the major reservoir for B. burgdorferi and, upon infection, certain inbred mice develop symptoms similar to those reported in human disease. Sonicated preparations of washed spirochetes were found to have potent mitogenic activity when cultured with lymphocytes from naive C57BL/6, C3H/HeJ, or BALB/c mice. The activity of the B. burgdorferi sonicate was approximately fourfold greater than that of a similarly prepared Escherichia coli sonicate. Polymyxin B efficiently inhibited the mitogenic activity of the E. coli sonicate but only slightly inhibited that of the B. burgdorferi sonicate, suggesting that a lipid A-containing lipopolysaccharide was not responsible for the B. burgdorferi activity. Kinetic analysis indicated peak proliferation at 2 to 3 days of culturing, suggesting polyclonal activation. B- and T-lymphocyte depletion experiments indicated that the major cell type responding to the B. burgdorferi mitogen was the B lymphocyte. This mitogen stimulated murine B cells not only to proliferate but also to differentiate into antibody-secreting cells, as demonstrated by the production of immunoglobulin by stimulated splenocytes. Furthermore, the sonicated preparation stimulated the B-cell tumor line CH12.LX to secrete immunoglobulin in the absence of accessory cells. B. burgdorferi also stimulated interleukin-6 production in splenocyte cultures. The observation that B. burgdorferi can stimulate activation of and immunoglobulin production by normal B lymphocytes may directly reflect on the development of arthritis associated with persistent infection by this organism.     

Involvement of birds in the epidemiology of the Lyme disease agent Borrelia burgdorferi.

Borrelia burgdorferi, the causative agent of Lyme disease, was isolated from the liver of a passerine bird, Catharus fuscescens (veery), and from larval Ixodes dammini (tick) feeding on Pheucticus ludovicianus (rose-breasted grosbeak) and Geothlypis trichas (common yellowthroat). In indirect immunofluorescence antibody tests, isolates reacted with polyclonal and monoclonal (H5332) antibodies. Studies on the DNA composition of the veery liver isolate and the strain cultured from an I. dammini larva indicated that both were B. burgdorferi and not Borrelia anserina or Borrelia hermsii. The veery liver isolate infected hamsters and a chick. In contrast, B. anserina infected chicks but not hamsters. B. burgdorferi is unique among Borrelia spp. in being infectious to both mammals and birds. We suggest that the cosmopolitan distribution of B. burgdorferi may be caused by long-distance dispersal of infected birds that serve as hosts for ticks.     

Infectious but nonpathogenic isolate of Borrelia burgdorferi.

We document for the first time an infectious but nonarthritogenic variant of Borrelia burgdorferi. Strain 25015, previously isolated from an Ixodes dammini larva collected in upstate New York, was infectious but failed to produce arthritis or carditis in laboratory rats and mice. By contrast, pathogenic strain N40 invariably caused arthritis. This nonarthritogenic variant, with proteins with molecular weights different from those of the standard B31 strain, was frequently isolated from normal joint tissues of experimentally infected rats. Outer surface proteins A and B of strain 25015 have molecular weights of about 32,500 and 35,500, respectively, in contrast to molecular weights of approximately 31,000 and 34,000, respectively, for outer surface proteins A and B of strains B31 and N40. A prominent low-molecular-weight protein of about 23,500 also characterizes strain 25015. Test animals infected for 30 to 60 days had relatively high antibody titers (greater than or equal to 1:1,280). The nonarthritogenic variant will be useful, along with pathogenic strains, in providing comparative insight into the pathogenesis of Lyme borreliosis. Homologous immunoblotting of sera from rats and mice inoculated with both the arthritogenic and nonarthritogenic strains revealed antibody reactivities to proteins of B. burgdorferi different from those revealed in the heterologous tests.     

Cystitis induced by infection with the Lyme disease spirochete, Borrelia burgdorferi, in mice.

Previous studies have demonstrated that the urinary bladder is a consistent source for isolating the Lyme disease spirochete, Borrelia burgdorferi, from both experimentally infected and naturally exposed rodents. We examined histopathologic changes in the urinary bladder of different types of rodents experimentally infected with Lyme spirochetes, including BALB/c mice (Mus musculus), nude mice (M. musculus), white-footed mice (Peromyscus leucopus), and grasshopper mice (Onychomys leucogaster). Animals were inoculated intraperitoneally, subcutaneously, or intranasally with low-passaged spirochetes, high-passaged spirochetes, or phosphate-buffered saline. At various times after inoculation, animals were killed and approximately one-half of each urinary bladder and kidney were cultured separately in BSK-II medium while the other half of each organ was prepared for histologic examination. Spirochetes were cultured from the urinary bladder of all 35 mice inoculated with low-passaged spirochetes while we were unable to isolate spirochetes from any kidneys of the same mice. The pathologic changes observed most frequently in the urinary bladder of the infected mice were the presence of lymphoid aggregates, vascular changes, including an increase in the number of vessels and thickening of the vessel walls, and perivascular infiltrates. Our results demonstrate that nearly all individuals (93%) of the four types of mice examined had a cystitis associated with spirochetal infection.     

Characterization of a circular plasmid from Borrelia burgdorferi, etiologic agent of Lyme disease.

Borrelia burgdorferi, the etiologic agent of Lyme disease, was recently shown to contain plasmid DNA. Two plasmid species have been described in strain CT1, a Wisconsin tick isolate: a 9.2-kilobase entity; and a larger, 70-kilobase entity. Characterization of the 9.2-kilobase entity by using DNase I and restriction endonucleases demonstrated that the plasmid is supercoiled and exists as a stable dimer in this strain. The role played by the plasmid in B. burgdorferi is unknown.     

Ticks and biting insects infected with the etiologic agent of Lyme disease, Borrelia burgdorferi.

Members of 18 species of ticks, mosquitoes, horse flies, and deer flies were collected in southeastern Connecticut and tested by indirect fluorescent-antibody staining methods for Borrelia burgdorferi, the etiologic agent of Lyme disease. An infection rate of 36.2% (116 tested), recorded for immature Ixodes dammini, exceeded positivity values for all other arthropod species. Prevalence of infection for hematophagous insects ranged from 2.9% of 105 Hybomitra lasiophthalma to 14.3% of seven Hybomitra epistates. Infected I. dammini larvae and nymphs coexisted with infected Dermacentor variabilis (American dog tick) immatures on white-footed mice (Peromyscus leucopus), but unlike I. dammini, none of the 55 adult American dog ticks collected from vegetation harbored B. burgdorferi. Groups of 113 field-collected mosquitoes of Aedes canadensis and 43 Aedes stimulans were placed in cages with uninfected Syrian hamsters. Of these, 11 females of both species contained B. burgdorferi and had fed fully or partially from the hamsters. No spirochetes were isolated from the hamsters, but antibodies were produced in one test animal.     

Measurement of antibodies to the Borrelia burgdorferi flagellum improves serodiagnosis in Lyme disease.

The isolation of Borrelia burgdorferi flagella and an enzyme-linked immunosorbent assay (ELISA) for detection of immunoglobulin G (IgG) and IgM to the B. burgdorferi flagellum are described. The diagnostic performance of the flagellum ELISA for serodiagnosis of Lyme disease was compared with the performance of a traditional whole cell B. burgdorferi sonic extract ELISA. We examined sera and cerebrospinal fluid (CSF) from 56 patients with lymphocytic meningoradiculitis (Bannwarth's syndrome), the most frequent secondary-stage manifestation of Lyme disease in Europe. Two hundred healthy individuals and patients with aseptic meningitis, encephalitis, Guillain-Barré syndrome, and syphilis served as controls. The flagellum ELISA was significantly more sensitive than the sonic extract ELISA. The diagnostic sensitivities were increased from 41.1 to 76.8% (P less than 0.01) for IgG and from 35.7 to 67.9% (P less than 0.05) for IgM detection in serum. The increase in sensitivity was most pronounced in patients with a short duration of disease (less than 20 days after onset). The diagnostic specificity increased for IgG detection but was almost unaltered for IgM. The flagellum ELISA did not improve the diagnostic sensitivity of measuring antibodies to borreliae in CSF, most likely owing to the low level of unspecific antibodies in CSF compared with serum. The cross-reactivity of sera and CSF from patients with syphilis decreased significantly. The flagellum antigen of B. burgdorferi shows no strain variation, is easy to purify in sufficient quantity, and is therefore a suitable reference antigen for routine serodiagnosis of Lyme disease.     

Ultrastructure of Borrelia burgdorferi in tissues of patients with Lyme disease

Spirochetal organisms were sought in 18 skin and 4 synovial membrane specimens obtained by biopsy from 22 Lyme disease patients. The presence of spirochetes in body tissues was histologically demonstrated in one patient with lymphadenosis benigna cutis, one patient with acrodermatitis chronica atrophicans and in one patient with active arthritis. The organisms were 5-30 microns long and 0.12-0.25 microns thick, had 8 or 11 flagella arising from both ends of the body, and their ultrastructure was analogous to that of cultured Borrelia burgdorferi strains. They were located intra- or perivascularly, or in the collagenous connective tissue of the skin and synovium. This implies that Lyme spirochetes may have a potential to survive in body tissues and cause injury to blood vessels.     

Reassessment of a midwestern Lyme disease focus for Borrelia burgdorferi and the human granulocytic ehrlichiosis agent

Previous studies from the late 1980s defined the risk of human Lyme disease by determining the prevalence of Borrelia burgdorferi infection in Ixodes scapularis ticks and Peromyscus sp. mice captured from areas around La Crosse, Wis. High percentages of B. burgdorferi-infected I. scapularis ticks and P. leucopus mice were common in areas located north of Interstate 90 but were not detected in areas south of this major east-west thoroughfare. In this study, we reevaluated the extent of B. burgdorferi infection. High percentages of mice captured from sites north of the interstate were still infected with B. burgdorferi. In addition, B. burgdorferi was recovered from 12 (67%) of 18 mice captured from a site well south of the highway. However, none of 104 mice or 713 I. scapularis ticks captured from the study sites were infected with Ehrlichia spp. The results confirmed the continued high risk for humans to contract infection with B. burgdorferi and the significant southward expansion of the area in which Lyme disease is endemic. In contrast, the risk of acquiring human granulocytic ehrlichiosis remains minimal despite the abundance of appropriate vector ticks and reservoir rodents.     

Serodiagnosis of Lyme disease by ELISA using Borrelia burgdorferi flagellum antigen

Antibodies to a 41,000 (41 kD) polypeptide in flagella of Borrelia burgdorferi were measured in patients with Lyme disease in Japan by flagellum ELISA. The IgG and IgM Classes of antibodies to a flagellum antigens were detected in the sera as early as 0.5 months after infection. The IgG antibodies continued to exist in their sera for more than one year, while the IgM antibodies quickly faded out from their sera. With respect to a diagnostic specificity of the flagellum ELISA, false positive reactions showing more than 10% were observed in sera with high levels of IgG or IgM, and with anti-syphilis antibody. This method, however, was unaffected by sera with high levels of IgA, rheumatoid factor or anti-nuclear antibody. In three cases of patients with erythema migrans preceded by tick-bite, and treated with antibiotics, seronegative results were observed by a immunoperoxidase (IP) test. Since two of them showed the positive level of IgM antibody by the flagellum ELISA, this method seems to be more sensitive and useful than the IP test for serodiagnosis of the Lyme disease.