Analysis of promoters in Borrelia burgdorferi by use of a transiently expressed reporter gene

A transient chloramphenicol acetyltransferase (CAT) expression system was developed for Borrelia burgdorferi. An Escherichia coli vector containing a promoterless Streptococcus agalactiae cat gene was constructed. Promoters for ospA, ospC, and flaB were placed upstream of this cat gene, and CAT assays were performed in E. coli from these stably maintained plasmids. The plasmids with putative promoters ospA and flaB were found to be approximately 20-fold more active than were the plasmids with ospC or no promoter. The level of activity correlated well with the resistance to chloramphenicol that each plasmid provided. Next, the nonreplicative plasmid constructs were transformed by electroporation into B. burgdorferi. CAT assays were performed by both thin-layer chromatography and the fluor diffusion method. Measurement of CAT activity demonstrated that the ospA promoter was again about 20-fold more active than the promoterless cat gene. The flaB and ospC promoters increased the activity seven- and threefold, respectively, over that with the promoterless construct. This simple transient-expression assay was shown to be an effective method to study promoter function in B. burgdorferi in the absence of a well-developed genetic system.

Immunologic and genetic analyses of VmpA of a neurotropic strain of Borrelia turicatae

In mice infected with serotype A but not serotype B of the relapsing fever spirochete Borrelia turicatae, early invasion of the brain occurs. Serotypes A and B are further distinguished by the abundant surface protein they produce: VmpA and VmpB, respectively. Western blotting with monoclonal antibodies, one-dimensional peptide mapping, and partial amino acid sequencing demonstrated regions of the VmpA protein that differed from VmpB. Oligonucleotide primers based on the partial amino acid sequences of unique regions were used to amplify a portion of the VmpA gene (vmpA) by PCR, and the product was used as a probe in Southern blot and Northern blot analyses. These experiments showed that (i) expression of the vmpA sequence was determined at the level of transcription and (ii) the vmpA sequence was in two locations in serotype A and one location in serotype B. The vmpA gene at the expression-linked locus of serotype A was cloned and sequenced. An open reading frame would encode a polypeptide of 214 amino acids. The polypeptide expressed by Escherichia coli was bound by VmA-specific but not VmpB-specific antibody. Primer extension analysis identified a consensus sigma70-type promoter for vmpA at the expression locus. Phylogenetic analysis revealed that VmpA is homologous to small Vmp (Vsp) proteins of B. hermsii and to OspC proteins of B. burgdorferi. These findings indicate that a function of the Vsp-OspC family of proteins of Borrelia spp. may be differential localization in organs, including the brain, during infection.

Antigenic variation in Lyme disease borreliae by promiscuous recombination of VMP-like sequence cassettes

We have identified and characterized an elaborate genetic system in the Lyme disease spirochete Borrelia burgdorferi that promotes extensive antigenic variation of a surface-exposed lipoprotein, VlsE. A 28 kb linear plasmid of B. burgdorferi B31 (lp28-1) was found to contain a vmp-like sequence (vls) locus that closely resembles the variable major protein (vmp) system for antigenic variation of relapsing fever organisms. Portions of several of the 15 nonexpressed (silent) vls cassette sequences located upstream of vlsE recombined into the central vlsE cassette region during infection of C3H/HeN mice, resulting in antigenic variation of the expressed lipoprotein. This combinatorial variation could potentially produce millions of antigenic variants in the mammalian host.

Oral delivery of purified lipoprotein OspA protects mice from systemic infection with Borrelia burgdorferi

The lipoprotein outer surface protein A (OspA) of the Lyme disease agent. Borrelia burgdorferi, has provided protection to mice and other animals against systemic infection when delivered orally as a recombinant protein in Escherichia coli, bacille Calmette. Guerin or Salmonella typhimurium. In the present study purified recombinant strain B31 OspA or outer surface protein D (OspD), another lipoprotein of B. burgdorferi, were administered either subcutaneously (s.c.) or orally without cell carrier or adjuvant to mice. In comparison to the OspD preparation, the OspA protein was 256-fold more resistant to trypsin. Whereas OspA in the suspension was in regular complexes of 17-25 nm in size, OspD formed amorphous globules of different sizes. Animals received a primary immunization and at least one booster. Mice immunized s.c. with either OspA or OspD had detectable antibodies to B. burgdorferi by enzyme-linked immunosorbent assay (ELISA), growth inhibition assay (GIA) and immunoblot. Delivered orally, OspA but not OspD elicited a specific antibody response, including IgA, as determined by these assays. The geometric mean titre of sera from mice who received 4 micrograms of OspA orally on days 1, 2, 4, 21 and 22 was 1470 by Ig ELISA, 320 by IgA ELISA and 128 by GIA. In infectious challenge experiments with B. burgdorferi strain Sh2-2-82 (OspA+ OspD- ) inoculated intradermally at 100 x the ID 50 all eight mice immunized with the 4 micrograms dose of OspA were protected, none of the mice immunized with the 4 micrograms dose of OspD were protected (P < 0.001 by Fisher exact test). These studies indicate that the lipoprotein OspA provides protection against systemic B. burgdorferi infection when delivered orally as a purified protein.

An OspA-based DNA vaccine protects mice against infection with Borrelia burgdorferi

Immunization with recombinant OspA protein of Borrelia burgdorferi protects against experimental Lyme disease. In the present study, mice were injected intramuscularly with plasmid DNA (VR2210) encoding strain B31 OspA. In this vector, the ospA-coding sequence was under transcriptional control of the cytomegalovirus immediate early promoter. For negative and positive controls, mice were immunized with either the plasmid vector without an osp-coding sequence or recombinant OspA protein, respectively. Mice immunized with VR2210 DNA produced OspA-specific antibodies that bound to B. burgdorferi in a whole cell ELISA and inhibited the growth of a homologous strain of B. burgdorferi. Immunization with VR2210 protected mice against challenge with 2 infectious strains of B. burgdorferi, Sh-2-82 and N40. These results indicate that vaccination with plasmid DNA expressing OspA is an efficacious method for providing a protective response against B. burgdorferi infection.

Arthritis severity and spirochete burden are determined by serotype in the Borrelia turicatae-mouse model of Lyme disease

Immunodeficient mice infected with Borrelia turicatae, a relapsing fever agent, have a disorder that resembles disseminated Lyme disease. Two serotypes, A and B, differed in their arthritogenicity in both CB-17 SCID and C3H SCID mice. In CB-17 SCID mice infected with serotype A or B, arthritis was assessed by measurement of tibiotarsal diameter, functional ability on a beam walk test, and microscopic assessment of joint inflammation. Serotype B-infected mice had greater joint swelling, functional disability, and leukocytic infiltration in the joints than serotype A-infected mice. Joint swelling and disability peaked at 2 weeks of infection and then decreased, while leukocyte infiltration in the joints persisted. To investigate the basis for the differences in arthritogenicity of serotypes A and B, spirochete burdens in infected mice were measured by quantitative PCR of spirochete DNA in joints, direct immunofluorescence of spirochetes in joints, and counts of spirochetes in the blood. At 2 weeks of infection there were seven times more spirochetes in the joints of serotype B-infected mice than in those of serotype A-infected mice, measured by both quantitative PCR and direct enumeration. Although serotypes A and B had the same infectivity and growth rate in vivo, serotype B spirochetes were eightfold more abundant in the blood than serotype A spirochetes and produced greater fatality in newborn mice. These findings indicate that differences in disease severity in mice infected with serotype A or B are attributable to differences in the spirochete burden in the joints and blood.

Surface exposure and species specificity of an immunoreactive domain of a 66-kilodalton outer membrane protein (P66) of the Borrelia spp. that cause Lyme disease

A chromosomally encoded 66-kDa protein (P66) of Borrelia spp. that cause Lyme disease has previously been shown to be associated with the spirochetal outer membrane. A topological model of P66 predicts a surface-exposed fragment which links the N- and C-terminal intramembranous domains of the protein (J. Bunikis, L. Noppa, and S. Bergström, FEMS Microbiol. Lett. 131:139-145, 1995). In the present study, an immunogenic determinant of P66 was identified by a comparison of the immunoreactivities of different fragments of P66 generated either by proteolytic treatment of intact spirochetes or as recombinant proteins expressed in Escherichia coli. The immune response to P66 during natural infection was found to be directed against the predicted surface domain which comprises amino acids at positions 454 through 491. A sequence comparison revealed considerable polymorphism of the surface domains of P66 proteins of different Lyme disease-causing Borrelia species. Five sequence patterns of this domain were observed in the B. garinii strains studied. In contrast, sequences of the relevant part of P66 of the B. afzelii and B. burgdorferi sensu stricto isolates studied were identical within the respective species. In immunoblotting, 5 of 17 (29.4%) sera from North American patients with early disseminated or persistent Lyme disease reacted against P66 of B. burgdorferi sensu stricto B31. These sera, however, failed to recognize P66 of B. afzelii and B. garinii, as well as an analog of P66 in the relapsing fever agent, B. hermsii. In conclusion, the topological model of P66 is supported by the demonstration of an apparent surface localization of an immunoreactive domain of this protein. Furthermore, analogous to the plasmid-encoded borrelial outer surface proteins, the predicted surface-exposed portion of chromosomally encoded P66 appears to be antigenically heterogenous.

In vivo activities of ceftriaxone and vancomycin against Borrelia spp. in the mouse brain and other sites

Borrelia burgdorferi, the agent of Lyme disease, and B. turicatae, a neurotropic agent of relapsing fever, are susceptible to vancomycin in vitro, with an MIC of 0.5 microgram/ml. To determine the activity of vancomycin in vivo, particularly in the brain, we infected adult immunocompetent BALB/c and immunodeficient CB-17 scid mice with B. burgdorferi or B. turicatae. The mice were then treated with vancomycin, ceftriaxone as a positive control, or normal saline as a negative control. The effectiveness of treatment was assessed by cultures of blood and brain and other tissues. Ceftriaxone at a dose of 25 mg/kg of body weight administered every 12 h for 7 to 10 days eliminated cultivable B. burgdorferi or B. turicatae from all BALB/c or scid mice in the study. Vancomycin at 30 mg/kg administered every 12 h was effective in eliminating infection from immunodeficient mice if treatment was started within 3 days of the onset of infection. If treatment with vancomycin was delayed for 7 days or more, vancomycin failed to eradicate infection with B. burgdorferi or B. turicatae from immunodeficient mice. The failure of vancomycin in eradicating established infections in immunodeficient mice was associated with the persistence of viable spirochetes in the brain during antibiotic treatment.

The nucleotide sequence of a linear plasmid of Borrelia burgdorferi reveals similarities to those of circular plasmids of other prokaryotes

A linear plasmid of Borrelia burgdorferi had 16,927 bp, a G+C content of 23.1%, a relative deficiency of CpG dinucleotides, and open reading frames A to O. The OrfC and OrfE proteins were similar to hypothetical proteins encoded by circular plasmids of B. burgdorferi. The OrfM and OrfN proteins were similar to replication proteins of circular plasmids of other bacteria.

A flagella-less mutant of Borrelia burgdorferi as a live attenuated vaccine in the murine model of Lyme disease

The immunogenicity and protective efficacy of an attenuated isolate of Borrelia burgdorferi, the agent of Lyme disease, were evaluated. An isogenic pair of isolates of strain HB19 differed in the expression of flagella; neither produced systemic infection or persisted in the skin of mice. In a comparison of intradermally administered live flagella-bearing and flagella-less cells, the flagella-less isolate was equal to or superior to the flagella-bearing wild type in eliciting growth-inhibiting antibodies and preventing infection in mice. In a comparison of live and killed flagella-less cells, immune responses to live cells were superior to those to killed cells, as assessed by ELISA, growth inhibition assay, and infectious challenge. The dose protecting 50% of mice was 10(3.8) and 10(5.2) for live and killed cells, respectively (P < .05).

Experimental immunization against Lyme borreliosis with recombinant Osp proteins: an overview

Interest in human and veterinary vaccines against Lyme borreliosis is growing. Both whole cell immunization and subunit vaccines can protect against infection with Borrelia burgdorferi. For development of a human vaccine the focus has been on a subunit vaccine. The most promising candidate is OspA, a major outer membrane lipoprotein of B. burgdorferi sensu lato. Of Osp proteins A through D, OspA shows the least variability between strains in its sequence and in the level of its expression. Borreliae in ticks express OspA. Antibodies to OspA kill borreliae in vitro and provide passive protection in mice. Active immunization of mice with OspA provides protection against challenge by syringe inoculation or tick bite. The lipid moiety of the OspA is necessary for immunogenicity in the absence of a potent adjuvant. A recombinant OspA-based vaccine is already in clinical trials. Although there is compelling evidence that immunization with OspA will provide protection, questions remain regarding the duration of protection from such immunization, the necessity to have a minimum level of neutralizing antibodies at all times for protection, and the relationship of an immune response to OspA and autoimmune features of Lyme borreliosis. The experimental aspects of immunization with Osp-A based constructs and other Lyme vaccine candidates are reviewed and discussed.

Emerging bacterial zoonotic and vector-borne diseases. Ecological and epidemiological factors

Among the etiologic agents of emerging infectious diseases are several bacterial organisms that naturally reside in animal and arthropod hosts. The most compelling emerging bacterial zoonotic and vector-borne diseases in the United States are Lyme disease; a Southern erythema migrans-like illness; human monocytic ehrlichiosis; human granulocytic ehrlichiosis; a novel cat flea-associated typhus group rickettsiosis; bartonelloses of immunocompetent and immunocompromised persons, particularly with AIDS; and sylvatic plague. Some of these antimicrobial-treatable infections are life threatening. During the acute stage of illness when antimicrobial agents are most effective, the flulike clinical signs and symptoms and available laboratory tests frequently do not point to a particular diagnosis. Epidemiological factors determined by the ecology of the bacteria are often the most useful diagnostic clues. The recognition of these evolving problems emphasizes the need for development of better laboratory diagnostic methods, for surveillance for and tracking of disease, and for continued research into factors contributing to transmission of the organisms. The continual appearance of previously unidentified bacterial infections requires prospective national strategies for timely recognition of the syndrome, identification of the agent, establishment of criteria and methods for diagnosis, optimization of the treatment regimen, and determination of successful approaches to prevention and control.

Conversion of a linear to a circular plasmid in the relapsing fever agent Borrelia hermsii

Spirochetes of the genus Borrelia have genomes composed of both linear and circular replicons. We characterized the genomic organization of B. burgdorferi, B. hermsii, B. turicatae, and B. anserina with pulsed-field gel electrophoresis. All four species contained a linear chromosome approximately 1 Mb in size and multiple linear plasmids in the 16- to 200-kb size range. Plasmids 180 and 170 kb in size, present in the relapsing fever agents B. hermsii and B. turicatae but not in the other two species, behaved as linear duplex DNA molecules under different electrophoretic conditions. A variant of strain HSI of B. hermsii had a 180-kb circular instead of linear plasmid. There were no detectable differences in the growth rates or in the expression of cellular proteins between cells bearing linear forms and those bearing circular forms of the plasmid. The conversion to a circular conformation of monomeric length was demonstrated by the introduction of strand breaks with irradiation, restriction endonuclease analysis, and direct observation of the DNA molecules by fluorescent microscopy. Consideration of different models for the replication of linear DNA suggests that circular intermediates may be involved in the replication of linear replicons in Borrelia spp.

Identification of an uncultivable Borrelia species in the hard tick Amblyomma americanum: possible agent of a Lyme disease-like illness

Bites from the hard tick Amblyomma americanum are associated with a Lyme disease-like illness in the southern United States. To identify possible etiologic agents for this disorder, A. americanum ticks were collected in Missouri, Texas, New Jersey, and New York and examined microscopically. Uncultivable spirochetes were present in approximately 2% of the ticks. Borrelia genus-specific oligonucleotides for the flagellin and 16S rRNA genes were used for amplification of DNA. Products were obtained from ticks containing spirochetes by microscopy but not from spirochete-negative ticks. Sequences of partial genes from spirochetes in Texas and New Jersey ticks differed by only 2 of 641 nucleotides for flagellin and 2 of 1336 nucleotides for 16S rRNA. Phylogenetic analysis showed that the spirochete was a Borrelia species distinct from previously characterized members of this genus, including Borrelia burgdorferi. Gene amplification could be used to detect these spirochetes in ticks and possible mammalian hosts.

Does Lyme disease occur in the south?: a survey of emerging tick-borne infections in the region

Lyme disease is the most common arthropod-borne infection in the United States. However, the risk of infection varies widely by geographic region. In the South, Borrelia burgdorferi has been identified in ticks and small mammals, but transmission of the agent to humans has not been documented. The Lyme disease-like disorder reported from the region may have another etiology.

High- and low-infectivity phenotypes of clonal populations of in vitro-cultured Borrelia burgdorferi

Borrelias that cause Lyme disease lose the ability to infect and cause disease in laboratory animals following 10 to 16 passages of in vitro culture. In this study, clonal populations of the Sh-2-82 (Sh2) and B31 strains of Borrelia burgdorferi were isolated by subsurface plating on BSK-II agar plates and examined for infectivity in the C3H/HeN mouse model. Mice were injected intradermally with 10(5) B. burgdorferi organisms, and the tibiotarsal joint, heart, and bladder were cultured 2 to 4 weeks postinfection to determine whether viable organisms were present. Clones exhibited either a high-infectivity phenotype, in which cultures were consistently positive at all organ sites, or a low-infectivity phenotype, in which a low proportion of cultures were positive (5 of 40 in a representative experiment). In an Sh2 population that had undergone five in vitro passages, 7 of 10 clones were of the high-infectivity phenotype, and the remaining clones were of the low-infectivity phenotype. The proportion of high-infectivity clones decreased with continued in vitro passage, with only 1 of 10 clones exhibiting the high-infectivity phenotype after 10 passages and 0 of 10 clones yielding positive cultures after 20 passages. Representative high- and low-infectivity clones from passage 5 Sh2 cultures had 50% infectious doses of 1.8 x 10(2) and 1 x 10(5), respectively. Subclones consistently reflected the same infectivity phenotypes as those of the parent clones. The protein profiles and plasmid contents of the high- and low-infectivity clones were compared and exhibited few discernible differences. On the basis of these results, the loss of infectivity during in vitro culture results from the outgrowth of low-infectivity clones and begins to occur within the first five in vitro passages. Further examination of clonal populations may lead to the identification of genetic and protein factors important in the virulence and pathogenicity of Lyme disease borrelias.

Borrelia burgdorferi mutant lacking Osp: biological and immunological characterization

All Borrelia burgdorferi sensu lato isolates characterized to date have one or a combination of several major outer surface proteins (Osps). Mutants of B. burgdorferi lacking Osps were selected with polyclonal or monoclonal antibodies at a frequency of 10(-6) to 10(-5). One mutant that lacked OspA, -B, -C, and -D was further characterized. It was distinguished from the OspA+B+ cells by its (i) autoaggregation and slower growth rate, (ii) decreased plating efficiency on solid medium, (iii) serum and complement sensitivity, and (iv) diminished capacity to adhere to human umbilical vein endothelial cells. The Osp-less mutant was unable to evoke a detectable immune response after intradermal live cell immunization even though mutant survived in mouse skin for the same duration as wild-type cells. Polyclonal mouse serum raised against Osp-less cells inhibited growth of the mutant but not of wild-type cells, an indication that other antigens are present on the surface of the Osp-less mutant. Two types of monoclonal antibodies (MAbs) with growth-inhibiting properties for mutant cells were identified. The first type bound to a 13-kDa surface protein of B. burgdorferi sensu stricto and of B. afzelii. The MIC of the Fab fragment of one MAb of this type was 0.2 micrograms/ml. The second type of MAb to the Osp-less mutant did not bind to B. burgdorferi components by Western blotting (immunoblotting) but did not bind to unfixed, viable cells in immunofluorescence and growth inhibition assays. These studies revealed possible functions Osp proteins in borrelias, specifically serum resistance, and indicated that in the absence of Osp proteins, other antigens are expressed or become accessible at the cell surface.

Expression of the flagellin gene in Borrelia is controlled by an alternative sigma factor

The flagellin genes from six Borrelia species were cloned, sequenced and characterized at the molecular level. The flagellin genes of two relapsing fever Borrelia species, B. hermsii and B. crocidurae, three Lyme disease genomic species, B. burgdorferi, B. afzelii and B. garinii, and the avian borreliosis agent, B. anserina, were compared and showed an 85-93% sequence identity to each other. Comparison of the fla genes from the different Lyme borreliosis spirochaetes revealed that they were 94-99% identical. Nucleotide sequencing of the fla gene and primer extension on isolated mRNA from both B. hermsii (as transcribed in Escherichia coli) and B. burgdorferi (as transcribed in the natural host) identified the putative transcriptional start points, the ribosomebinding sites and the promoter regions of these genes. The deduced promoter of the Borrelia flagellin gene resembled neither the sigma 70 promoter of prokaryotes, as seen for the genes for the outer-surface proteins A and B in Lyme disease Borrelia and the genes for the variable major proteins 7 and 21 of B. hermsii, nor the sigma 28 consensus promoter region of motility genes from other bacteria. Instead, the promoter of the fla gene in Borrelia has most similarity to the bacteriophage SP01 sigma gp33-34 promoter sequence of Bacillus subtilis.

Antigen diversity in the bacterium B. hermsii through "somatic" mutations in rearranged vmp genes

B. hermsii counters host immunity with multiphasic antigenic variation. This is conferred by interplasmidic and intraplasmidic rearrangements of vmp genes. In several independent events, activation of a silent vmp gene through intraplasmidic deletions but not interplasmidic recombinations was followed by the appearance at its 5' end of multiple mutations that were not present in the silent gene. The prevalence of mutant alleles in postwitch populations increased during infections. Differences between the silent and expressed genes were at the same nucleotides at which vmp pseudogenes differed, suggesting these were templates for postswitch gene conversions. The mechanism of this bacterium to generate diversity, namely, intramolecular deletions followed by mutations in the rearranged gene, mirrors the strategy used by vertebrate hosts to eliminate it.

Activation of a vmp pseudogene in Borrelia hermsii: an alternate mechanism of antigenic variation during relapsing fever

The relapsing fever agent, Borrelia hermsii, undergoes multiphasic antigenic variation to evade its host's immune response. A frequently observed switch is serotype 7 to 26. Unlike silent vmp genes previously characterized, the transcriptionally silent vmp26 sequence was a pseudogene in lacking a start codon. In serotype 7 the location of the silent vmp26 sequence just downstream of vmp7 on the expression plasmid, as well as on the silent plasmid, was also unique. The demonstration of a predicted circular recombination product in serotype 7 but not serotype 21 populations indicates that the pseudogene was activated by an intramolecular recombination producing a deletion of DNA between 20-nucleotide direct repeats in vmp7 and psi vmp26.

A family of surface-exposed proteins of 20 kilodaltons in the genus Borrelia

Relapsing fever and Lyme disease spirochetes of the genus Borrelia display at their surfaces abundant lipoproteins: Vmp proteins in Borrelia hermsii and Osp proteins in Borrelia burgdorferi. Vmp and Osp proteins largely determine serotype specificity, and neutralizing antibodies of infected or immunized animals are directed at them. For the present study, we examined B. hermsii serotype 33, which is unique among strain HS1 serotypes in the low frequency of switches to other serotypes during infections and in vitro cultivation. Failing to clone the complete vmp33 gene, we accomplished its further characterization by (i) determining three partial amino acid sequences, (ii) designing oligonucleotide primers based on these amino acid sequences, (iii) cloning and sequencing the central portion of vmp33, and (iv) using outwardly directed primers and the inverse PCR to clone the 5' and 3' ends of the gene and flanking regions. The transcriptional start site was identified by primer extension analysis. Vmp33 was a polypeptide of 211 amino acids; the three partial amino acid sequences were identified in the open reading frame. Vmp33 was found to be more similar to other 20-kDa Vmp proteins of B. hermsii and to OspC proteins of B. burgdorferi than it was to 35- to 39-kDa Vmp proteins of the same strain. Moreover, OspC proteins were more similar to Vmp33 than they were to OspA, -B, or -D proteins of B. burgdorferi. These sequence similarities were consistent with Western blot (immunoblot) findings of cross-reactions between Vmp33 and OspC with anti-Vmp33 and anti-OspC sera. The promoter for the expressed vmp33 gene was found to be different from the expression site for other active vmp genes characterized to date. These results indicate that Vmp33 and other small Vmp's belong with OspC to a genus-wide family of 20-kDa proteins and that expression of these proteins may be coordinated with expression of other Vmp and Osp proteins in Borrelia spp.