Search

Borrelia burgdorferi complement regulator-acquiring surface protein 1 (CRASP-1), the dominant factor H and FHL-1-binding protein of the Lyme disease spirochete B. burgdorferi, is implicated in pathogen persistence and was recently reported to be nonimmunogenic in humans. Here we show that serum samples from Lyme disease patients contain antibodies with exclusive specificity for nondenatured structural determinants of CRASP-1.

Complement regulator-acquiring surface protein 1 (CRASP-1) is the dominant factor-H-like protein 1 (FHL-1)- and factor-H-binding protein of Borrelia burgdorferi and is suggested to contribute to persistence of the pathogen. The prototype CRASP-1 of B. burgdorferi sensu stricto (CRASP-1Bb) has been formerly characterized. As shown recently, serum-resistant Borrelia afzelii strains express a unique FHL-1 and factor H-binding protein, designated CRASP-1Ba. Here, we describe for the first time the isolation and functional characterization of the gene encoding the full-length CRASP-1Ba of 28 kDa, which, upon processing, is predicted to be 26.4 kDa. CPASP-1Ba of B. afzelii spirochetes is associated with a genetic locus encoding the orthologous gbb54 gene family that maps to the linear plasmid of approximately 54 kb. Ligand affinity blotting techniques demonstrate that both native and recombinant CRASP-1Ba molecules strongly bind to FHL-1 and much more weakly to factor H. The FHL-1 and factor-H-binding site in CRASP-1Ba is shown to be localized to a 12-amino-acid residue domain at the C terminus of the protein. For comparison, the corresponding cspA-like gene(s) of a serum-sensitive Borrelia garinii strain has also been cloned and characterized. Most notably, two CRASP-1-related B. garinii proteins were identified; however, both molecules bind only weakly to FHL-1 and not at all to factor H. The present identification of the binding site of CRASP-1Ba represents an important step forward in our understanding of the pathogenesis of Lyme disease and may be helpful to design therapeutic regimens to interfere with complement evasion strategies of human pathogenic Borrelia strains.

DNA vaccines encoding the outer surface protein A (OspA) of Borrelia burgdorferi have been shown to induce protective humoral responses capable of preventing but not curing infection in mice. Subsequent studies showed that an established infection or disease could be resolved by passive transfer of antibodies to OspC. In the present study, DNA vaccines encoding either the OspC antigen alone or fused to OspA and under the transcriptional control of the human elongation factor 1alpha promoter were evaluated for their protective and/or curative potential. In contrast to ospA-containing plasmids, none of the six constructs with ospC alone were immunogenic in vivo, independent of whether they contained promoter or leader sequences from ospA and/or ospC, or alternatively, the signal sequence of the human tissue plasminogen activator. Solely, a DNA vaccine encoding an OspA-OspC fusion product led to expression of the respective polypeptide chain in transfected cells in vitro and to the induction of OspA- and OspC-specific antibodies in vivo. Immune sera raised against the OspA-OspC fusion product conveyed full protection against subsequent infection, most probably via OspA-specific antibodies, but were unable to resolve infection.

Borrelia burgdorferi is transmitted by infected ticks and causes Lyme disease. To infect distant organ sites, B. burgdorferi spirochetes must disseminate from the site of the tick bite. During dissemination from the dermal tissue, they breach tissue barriers, probably by proteolysis. The previous findings that spirochetes bind serum-derived plasminogen and that plasmin favors spirochetal invasiveness and infectivity suggested a role for plasmin in the pathogenicity of B. burgdorferi. Binding of plasminogen to spirochetes and activation into plasmin is favored in a microenvironment that is rich in plasminogen and plasminogen activators. Plasminogen is abundant in plasma and interstitial fluids, and it is increased in inflammatory exudates. Since B. burgdorferi does not express endogenous plasminogen activators, the conversion of spirochete-bound plasminogen depends on host-derived plasminogen activators. In this report, we show that both intact B. burgdorferi organisms and its recombinant outer surface lipoprotein A induce human monocytes to express and secrete urokinase-type plasminogen activator in its zymogen form (pro-uPA). Moreover, we demonstrate that the presence of B. burgdorferi accelerates the interaction between (pro-)uPA and plasmin(ogen), leading to spirochete-bound plasmin. In a pro-uPA-serum mixture, spirochete-bound plasmin activity is generated. Taken together, the data suggest that B. burgdorferi may induce pro-uPA in a monocyte-containing inflammatory site and that the spirochetal surface provides an appropriate milieu for subsequent interactions between (pro-)uPA and plasmin(ogen), which result in spirochete-bound plasmin even in the presence of inhibitors for plasminogen activators and plasmin.

Previously we have found that sera from immunocompetent mice infected either naturally by ticks or experimentally with low numbers of Borrelia burgdorferi ZS7 bacteria lack OspA- and OspB-specific antibodies but confer optimal protection on severe combined immunodeficiency mice against challenge with spirochetes (U.E. Schaible, L. Gern, R. Wallich, M. D. Kramer, M. Prester, and M. M. Simon, Immunol. Lett. 36:219-226, 1993). We have now used the latter immune sera to identify new spirochetal structures with relevance for protection from an expression library of the virulent European strain B. burgdorferi ZS7. Here we report the cloning and characterization of a novel lipoprotein, designated pG, the gene for which is located on a 48-kb linear plasmid. Sequence analysis of the pG gene revealed an open reading frame encoding a putative lipoprotein of 196 amino acids with a calculated molecular mass of 22 kDa and a consensus cleavage sequence (Leu-X-Y-Z-Cys) recognized by signal peptidase II. Restriction fragment length polymorphism analyses of pG derived from independent B. burgdorferi isolates from different geographic areas revealed that the gene is species specific, with, however, extensive genotypic heterogeneity. Comparison of the protein sequence of pG with those of other known B. burgdorferi outer surface lipoproteins (OspA to OspF and P27) demonstrated that pG is most related to OspF. Furthermore, the upstream region of pG exhibited extensive sequence homology (> 94%) with the ospEF promoter region. Mouse immune sera to recombinant pG did not recognize a corresponding molecule in lysates of in vitro-propagated ZS7 spirochetes. However, experimental or natural infection of mice with ZS7 resulted in the induction of antibodies with reactivity for pG and the potential to delay the development of clinical arthritis. Together with the finding that sera from Lyme disease patients also contain antibodies to pG, our data suggest that the pG gene is preferentially expressed in the mammal environment.

Immune responses to Borrelia burgdorferi and their influence on spirochete transmission to Ixodes ricinus were analyzed in the natural European reservoir hosts; i.e., the mouse species Apodemus flavicollis (yellow-necked mouse) and Apodemus sylvaticus (wood mouse) and the vole species Clethrionomys glareolus (bank vole), and, in addition, in the laboratory mouse strain NMRI. Naive and preimmunized rodents were infected either by artificially infected I. ricinus larvae or by intradermal injection of spirochetes. Independent of the species, all animals developed antibodies to various spirochetal antigens. However, antibodies to the outer surface proteins A (OspA) and B (OspB) were not found in recipients infected via ticks. Rodents of the genus Apodemus and of the NMRI strain showed higher levels of B. burgdorferi-specific antibodies than those of the species C. glareolus. The rate of spirochete transmission to noninfected ticks correlated with both the quality and quantity of spirochete-specific antibodies generated in the various species: high levels of spirochete-specific immunoglobulins correlated with low transmission rates. Furthermore, lower transmission rates were observed with rodents expressing antibodies to OspA and OspB (i.e., intradermally infected or immunized) than with those lacking these specificities (i.e., infected via ticks). The study provides evidence that transmission of B. burgdorferi from natural hosts to ticks is controlled by the specificity and quantity of spirochete-reactive antibodies and suggests that immunity to B. burgdorferi in natural reservoir hosts is an important regulatory factor in the horizontal transmission of B. burgdorferi in nature.

We describe the cloning, expression, and molecular characterization of a novel polymorphic Borrelia burgdorferi lipoprotein recognized by monoclonal antibody LA7. Sequence analysis revealed an open reading frame encoding a 21,866-Da polypeptide (IpLA7). Comparison with other known proteins indicated sequence similarity between IpLA7 signal peptides and those of other prokaryotic lipoproteins, including the immunodominant B. burgdorferi outer surface proteins OspA, OspB, pC, and OspD. Both natural IpLA-7 and recombinant IpLA-7 could be biosynthetically labeled with [3H]palmitate. Upon solubilization of intact B. burgdorferi with the nonionic detergent Triton X-114, IpLA7 was extracted together with other lipoproteins into the detergent phase. Indirect immunolabeling studies indicated that the epitope recognized by monoclonal antibody LA7 is mainly located in the periplasmic space. Two-dimensional gel electrophoresis and immunoblotting confirmed the calculated acidic pI of 5.7 for IpLA-7. The LA7 gene was shown to be species specific and to be located on the linear chromosome of B. burgdorferi. The analysis of 40 individual spirochetal isolates on the basis of restriction fragment length polymorphisms revealed considerable genotypic heterogeneity of LA7 corresponding to that previously found for ospA. Native IpLA-7 and recombinant IpLA-7 were recognized by immune sera from infected mice as well as some human sera derived from infected but healthy donors and may thus prove useful as an additional marker for the serodiagnosis of Lyme disease.

In order to assess the genetic variation of immunologically relevant structures among isolates of the Lyme disease spirochete, Borrelia burgdorferi, three chromosomal genes encoding flagellin (fla) and the heat shock proteins HSP60 and HSP70, as well as the plasmid gene encoding outer surface protein A (OspA), from 55 different European and North American strains obtained from ticks and mammal hosts have been investigated by restriction fragment length polymorphisms (RFLPs). RFLPs of fla and the HSP60 and HSP70 genes revealed two distinct banding patterns (A and B) for each of the three genes and allowed the definition of four genomic groups [AAA, BBB, BBA, and B(A/B)A] for the three chromosomal genes. On the other hand, RFLPs of the OspA gene revealed six distinct banding patterns (types I to VI) making up six independent genomic groups for the plasmid-encoded gene. Furthermore, we have sequenced the chromosomal HSP60 gene from B. burgdorferi ZS7 and the plasmid-encoded OspA gene from two strains, ZQ1 and 19857. Alignment of the deduced HSP60 amino acid sequence from B. burgdorferi ZS7 (genomic group AAA) to a previously published HSP60 sequence derived from strain ACA-1, which according to the proposed classification is in a different genomic group (BBA), revealed a sequence identity of > 99%. Similar alignments of the OspA sequence of strain ZQ1 to those of other isolates that were published previously revealed sequence identities of between 70 and 94% among strains of distinct OspA genomic groups. These data indicate the existence of a restricted number of species-specific subgroups and clearly show that genotypic variation is much more pronounced for the OspA gene than for fla and the HSP60 and HSP70 genes. A phylogenetic tree constructed on the basis of distance matrix analyses of 12 OspA sequences supports the proposed classification of genomic groups of B. burgdorferi.

The uptake mechanism for the spirochete Borrelia burgdorferi, the causative agent of Lyme disease, was investigated by electron microscopy for human and murine phagocytes. Spirochetes of both a low- and a high-passage strain were preferentially internalized by coiling rather than by conventional phagocytosis. The spirochetes engulfed by coiling phagocytosis were found to disintegrate in an organelle exclusion zone without evident participation of lysosomes. Preincubation of B. burgdorferi with monoclonal antibody to the spirochetal OspA enhanced phagocytosis in general but did not consistently influence the uptake mechanism. Quantitative and kinetic differences concerning the phagocytic rate and mechanism were evident between cells from different lineages, different human individuals, and mice and humans. In general, when few phagocytes participated in spirochete uptake, the active cells displayed a high ratio of coiling versus conventional phagocytosis. These results suggest that coiling phagocytosis of B. burgdorferi plays a critical role in the control of spirochetal infection. More detailed studies on the molecular basis of this phagocytic mechanism may lead to new insights into the pathogenesis of Lyme borreliosis, a disease which is frequently characterized by the host's inability to eliminate the pathogenic spirochete.

The expression of T-cell-associated serine proteinase 1 (MTSP-1) in vivo during Leishmania major infection was analyzed in genetically resistant C57BL/6 mice and in genetically susceptible BALB/c mice. Using a monoclonal antibody as well as an RNA probe specific for MTSP-1 to stain tissue sections, we found T cells expressing MTSP-1 in skin lesions and spleens of mice of both strains. In skin lesions, MTSP-1-positive T cells could be detected as early as 3 days after infection. Most importantly, the frequency of T cells expressing MTSP-1 was significantly higher in susceptible BALB/c mice than in resistant C57BL/6 mice. These findings suggest that MTSP-1 is associated with disease-promoting T cells and that it may be an effector molecule involved in the pathogenesis of cutaneous leishmaniasis.

Monoclonal antibodies directed against the major Borrelia burgdorferi flagellar protein, the 41-kilodalton (kDa) protein flagellin, were used to monitor cloning and expression of the flagellin gene from a Borrelia burgdorferi genomic library. The structure of the gene was analyzed, and recombinant nonfusion flagellin was produced in Escherichia coli. A DNA sequence analysis of the 41-kDa flagellin gene revealed the presence of an open reading frame that encoded a protein having 336 amino acid residues and a calculated molecular mass of 35.8 kDa, indicating that there was posttranslational modification of the natural 41-kDa flagellin protein. Upstream from the AUG start codon sequence we identified motifs corresponding to consensus procaryotic promoter elements which could be utilized by the cloned flagellin gene when it was expressed in E. coli MC1061. The deduced flagellin protein sequence exhibited high levels of homology to sequences of flagellin proteins from Bacillus subtilis and Salmonella typhimurium. The levels of sequence similarity for the amino- and carboxy-terminal portions were about 65 and 56%, respectively. DNA sequence information on the flagellin gene was used to design oligonucleotides for gene amplification by the polymerase chain reaction method, and by using this method 0.01 pg of Borrelia burgdorferi DNA could be detected. Our results provide a basis for further biochemical analysis of the 41-kDa flagellin protein, investigation of the role of this protein in host-pathogen interactions, and development of a standardized reagent for diagnostic systems for Borrelia burgdorferi infections.

Peritoneal exudate T lymphocytes from mice experimentally infected with the intracellular bacterium Listeria monocytogenes secreted high interleukin 2 activities after interaction with syngeneic normal macrophage presenting listerial antigen in vitro. L. monocytogenes-immune cells secreting IL 2 were radioresistant and bore the phenotype Thy 1(+) Lyt 1(+)23(-).

Antigen-specific T-cell responses and histopathological changes were studied in mice experimentally inoculated with Borrelia burgdorferi B31. Inbred mice with different H-2 haplotypes and/or different genetic backgrounds were inoculated with B. burgdorferi organisms and tested for antigen-specific T-cell responses in vivo (delayed-type hypersensitivity [DTH]) and in vitro (T-cell proliferation). Comparable DTH responses were found after inoculation with either inactivated (in the presence of adjuvants) or viable microorganisms in all mouse strains, except BALB/c, irrespective of the H-2 haplotype (b, d, k, or s) tested and the sex of the animals. Moreover, in mice presensitized to B. burgdorferi, DTH responses could be induced only with antigen preparations derived from the corresponding strain but not with those obtained from either related spirochetes such as Treponema phagedenis and Leptospira interrogans or unrelated bacteria such as Mycobacterium tuberculosis. T cells isolated from lymph nodes or spleens of mice previously sensitized to B. burgdorferi but not those from naïve mice could be induced for antigen-specific proliferation in vitro, as revealed by [3H]thymidine incorporation. Histopathological examination of mice inoculated with viable B. burgdorferi organisms revealed significant perivascular infiltrates consisting mainly of mononuclear and a few polymorphonuclear leukocytes in different organs (brain, heart, lungs, liver, and kidneys) and the appearance of giant multinucleated cells within the spleen similar to those found in human skin specimens of patients suffering from cutaneous manifestations of Lyme disease. Our findings suggest that mice are a suitable animal model with which to study the immune response to B. burgdorferi and the pathogenesis of Lyme disease.

The role of splenic macrophages in resistance to lethal Plasmodium berghei or self-limiting Plasmodium yoelii was studied by testing their rate of phagocytosis and their production of O2 metabolites (H2O2 and O2-) upon nonspecific stimulation with zymosan. It was found that, compared with P. berghei, infection of mice with P. yoelii resulted in an earlier appearance and in higher numbers of adherent cells in the spleen. Furthermore, the capacity of macrophages to generate O2 metabolites was significantly higher in P. yoelii- than in P. berghei-infected mice. This difference in the production of O2 metabolites was more pronounced when calculated on a per spleen basis. On the other hand, phagocytosis by macrophages was similar in both types of infection. The data suggest that lethal and nonlethal malaria species differ in their capacity to induce the production of O2 metabolites by macrophages, thereby influencing the final course of disease.

Peritoneal exudate T lymphocytes from Listeria monocytogenes-immune mice in the presence of the homologous antigen (heat-killed L. monocytogenes) and normal macrophages showed L. monocytogenes-specific proliferative responses. Proliferation was inhibited by macrophages from L. monocytogenes- or Corynebacterium parvum-pretreated mice as well as by exogenous prostaglandin E(2). Macrophage-dependent inhibition of T-cell proliferation-at least in part-could be reversed by addition of indomethacin. When selected L. monocytogenes-immune Lyt T-cell subsets were cultured in the presence of inhibitory macrophages, pretreatment with anti-Lyt 1 antiserum plus complement completely abrogated proliferation and pretreatment with anti-Lyt 2 and anti-Lyt 3 antisera plus complement markedly reduced proliferation. However, a mixture (1:1) of the two preselected Lyt T-cell subsets resulted in complete reconstitution of proliferative responses. In contrast, when L. monocytogenes-immune peritoneal exudate T lymphocytes were treated with anti-Lyt antisera plus complement after culture, only treatment with anti-Lyt 1 antiserum plus complement affected proliferation, suggesting regulatory interactions between Lyt 1(+)23(-) and Lyt 1(-)23(+) T cells during in vitro culture which result in proliferation within the Lyt 1(+)23(-) T-cell subset. After rigorous depletion of residual macrophages and in the presence of indomethacin, pretreatment with anti-Lyt 1 antiserum plus complement, but not with anti-Lyt 2 and 3 antisera plus complement, eliminated proliferation. The data presented indicate that interactions between macrophages and Lyt T-cell subsets regulate L. monocytogenes-specific T-cell activation.

For analysis of the role of immune T cells in protective immunity against murine malaria, Plasmodium yoelii-immune Lyt T-cell subsets were functionally characterized in vitro and in vivo. Selected Lyt2- and Lyt2+ T cells from P. yoelii-immune C57BL/10 mice differed in their capability to proliferate in response to P. yoelii antigen in vitro. Only the Lyt2- T-cell population produced T-cell growth factor upon restimulation, and none of the selected T-cell subsets produced detectable amounts of macrophage activating factor. Lyt2- but not Lyt2+ lymphocytes were capable of transferring protection to normal C57BL/10 mice. When transferred into T-cell-deficient C57BL/6-nu/nu mice, adoptive resistance to P. yoelii by Lyt2- lymphocytes was only demonstrable after prior reconstitution of recipients with normal T cells. These results suggest an interaction between P. yoelii-immune Lyt2- T cells and normal T lymphocytes via T-cell growth factor in the development of protective immunity to malaria.