INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Lyme disease (borreliosis) was first recognized in juvenile patients from Lyme, Conn., in the late 1970s (45). Currently, Lyme borreliosis is the most frequently reported tick-associated illness in the United States, with over 13,000 cases reported in 1996 (9). Infection with Borrelia burgdorferi, the etiologic agent of Lyme borreliosis, is transmitted through the bite of several different species of ticks belonging to the genus Ixodes (5, 44). Classically, Lyme borreliosis begins with an expanding, ring-shaped skin lesion, erythema migrans, which usually develops at the site of the tick bite (41). However, since only 60 to 80% of infected individuals develop erythema migrans, infection with B. burgdorferi may remain undetected. If the illness is undiagnosed or improperly treated, spirochetes disseminate to multiple sites. Disseminated infection is often manifested by secondary annular skin lesions (42), meningitis, Bell's palsy, atrioventricular blockage (43), or migratory pain in joints, muscles, or bone (46). Late or persistent Lyme borreliosis begins months to years following infection and may consist of intermittent or chronic arthritis (46), neurologic abnormalities (27), acrodermatitis chronica atrophicans (2), or other complications. Furthermore, up to 25% of patients do not respond to therapy and lingering complications of the infection may persist (41).

The serious morbidity that results from infection with B. burgdorferi has intensified research efforts to develop a safe, effective, comprehensive vaccine. Vaccination with several outer surface proteins (Osps) of B. burgdorferi has been successful in inducing protection against infection with B. burgdorferi. These proteins include OspA (31 kDa) (12, 13, 40, 47), OspB (34 kDa) (14, 34), OspC (22-23 kDa) (17, 33, 34), and the 39-kDa protein (38). Of these vaccinogens, OspA has emerged as the leading Lyme borreliosis vaccine candidate. Several investigators (3, 8, 28-30, 34, 35) have shown that induction of borreliacidal antibody is responsible for the protection of OspA-vaccinated animals against infection with B. burgdorferi. We (32) also showed that significant borreliacidal antibody developed in humans and hamsters 60 days after primary and secondary vaccination with high concentrations of recombinant OspA. Unfortunately, the anti-OspA borreliacidal antibody response waned rapidly. The rapid waning of the protective borreliacidal antibody response needs to be prevented to ensure long-term protection. Infection with B. burgdorferi has occurred in one OspA-vaccinated human (37) and animals (15, 48).

In this study, we determined that T cells obtained from vaccinated hamsters influence the production of the protective borreliacidal antibody response. When T and B cells obtained from hamsters 14 days after vaccination, but not for 7 or 21 days, were incubated in vitro with macrophages and B. burgdorferi, considerable anti-OspA borreliacidal antibody was produced. Furthermore, T cells from hamsters 7 or 21 days after vaccination prevented the 14-day immune T and B cells from producing borreliacidal antibody. These studies provide evidence that immune T cells obtained 1 or 3 weeks after vaccination can prevent the development and maintenance of a sustained anti-OspA borreliacidal antibody response.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Hamsters. Six- to eight-week-old inbred LSH hamsters were obtained from our breeding colony at the Wisconsin State Laboratory of Hygiene. Hamsters weighing 100 to 140 g were housed three per cage at an ambient temperature of 21°C. Food and water were provided ad libitum.

Organism. Low-passage (<10) virulent B. burgdorferi sensu stricto isolate 297 was cultured once in modified Barbour-Stoenner-Kelly medium (BSK) (4, 6) at 32°C to a concentration of 5 × 10⁷ spirochetes per ml. Five-hundred-microliter samples were then dispensed into 1.5-ml screw-cap tubes (Sarstedt, Newton, N.C.) containing 500 µl of BSK supplemented with 30% glycerol (Sigma Chemical Co., St. Louis, Mo.), sealed, and stored at 70°C. When needed, a frozen sample of spirochetes was thawed and used to inoculate fresh BSK. The number of spirochetes was determined using a Petroff-Hausser counting chamber under dark-field microscopy.

Vaccine preparation. B. burgdorferi 297 organisms were grown in 1 liter of BSK to log phase, pelleted by centrifugation (10,000 × g, 15°C, 10 min), and washed three times with phosphate-buffered saline (PBS; pH 7.4). The washed pellet was resuspended in 1% formalin, incubated at 32°C for 30 min with periodic mixing, then washed three times by centrifugation (12,000 × g, 10°C, 15 min), and resuspended in PBS. The borrelial protein concentration was determined by using a commercially available kit (Sigma). To make the vaccine, B. burgdorferi 297 organisms were suspended in 10 ml of a 1% suspension of aluminum hydroxide (Imject alum; Pierce, Rockford, Ill.) at a concentration of 250 µg of borrelial protein per ml.

Vaccination of hamsters. Hamsters were mildly anesthetized with ether contained in a nose-and-mouth cup and vaccinated intramuscularly in each hind thigh with 0.2 ml of the vaccine suspension containing 50 µg of borrelial protein in alum. Nonvaccinated hamsters were included as controls.

Antibody reagents. Hybridoma cell line 14-4-4s (ATCC HB-32) secretes murine monoclonal antibody (MAb) that recognizes a cell surface marker on hamster B lymphocytes (25, 26, 31, 49, 50) and has been shown to successfully separate hamster T and B lymphocytes (25, 26, 50). Hybridoma 14-4-4s was grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 15% bovine calf serum (HyClone Laboratories, Inc., Logan, Utah) at 37°C in a humidified atmosphere of 7.5% CO₂. After 7 to 10 days, the culture supernatant was collected following centrifugation at 500 × g for 10 min at 4°C, dispensed into 12-ml aliquots, and frozen at 20°C until used. Subsequently, some of the aliquots were thawed and used at a final dilution of 1:10. For T- and B-lymphocyte panning and enrichment procedures, 100 µg of unconjugated goat anti-mouse immunoglobulin, heavy and light chain specific (Organon Teknika Corporation, Durham, N.C.), in coupling buffer (15 mM Na₂CO₃, 35 mM NaHCO₃ [pH 9.6]) was used to coat 100- by 20-mm tissue culture dishes (Corning Glass Works, Corning, N.Y.). Dishes were maintained overnight at 4°C and washed four times with PBS before the addition of suspensions of lymph node cells that had been incubated with MAb 14-4-4s. T- and B-lymphocyte preparations were stained before and after panning for the presence of B lymphocytes. Cell preparations were stained with a phycoerythrin-conjugated goat anti-hamster immunoglobulin specific for both heavy and light chains (Boehringer Mannheim Biochemicals, Indianapolis, Ind.) and were analyzed by flow cytometry for the presence of B lymphocytes. Phycoerythrin-conjugated goat and rat immunoglobulins were used as isotype controls.

Macrophage recovery. Nonvaccinated hamsters were mildly anesthetized with ether contained in a nose-and-mouth cup and injected intraperitoneally with 5 ml of 3% aged thioglycolate medium (Sigma) in PBS. Three days after injection, hamsters were euthanized by inhalation of CO₂. Twenty milliliters of cold Hanks balanced salt solution was injected intraperitoneally, the peritoneal cavity was massaged for 1 min, and the peritoneal exudate cells were recovered by aspiration with a syringe. The peritoneal exudate suspensions from several animals were pooled and centrifuged at 500 × g for 10 min at 4°C. The supernatants were decanted, and the cells were resuspended in DMEM supplemented with 10% heat-inactivated (56°C for 30 min) bovine calf serum (HyClone) and 5 × 10⁵ M 2-mercaptoethanol (Sigma). Following resuspension, the cells were poured over 100- by 22-mm polystyrene tissue culture dishes (Corning Glass Works) and incubated at 37°C in an atmosphere of 7.5% CO₂ for 4 h. After incubation, the tissue culture dishes were gently rinsed twice with 10 ml of warm Hanks balanced salt solution to remove nonadherent cells; 5 ml of cold, nonenzymatic cell dissociation solution (Sigma) was then added to the dish, which was incubated at 4°C for 30 min. Macrophages were subsequently removed by vigorously tapping and gently scraping the inside of the tissue culture dish with a sterile rubber policeman. Recovered macrophage suspensions were pooled and centrifuged at 500 × g for 10 min at 4°C. After centrifugation, the supernatant was decanted and the macrophages were resuspended in cold PBS. Macrophage viability was determined by trypan blue exclusion. Giemsa stained smears of the isolated cells showed a homogeneous population of macrophages with no other types of leukocytes visible.

Isolation and enrichment of T lymphocytes. Isolation and enrichment of T lymphocytes with MAb 14-4-4s was performed by procedures described previously (25, 26, 49, 50). Briefly, T lymphocytes were isolated from the inguinal lymph nodes of hamsters 7, 14, or 21 days after vaccination with formalin-inactivated B. burgdorferi 297 and from nonvaccinated hamsters. Single-cell suspensions of lymph node cells were prepared by teasing apart the lymph nodes with forceps and pressing them through a sterile stainless-steel 60-mesh screen into DMEM containing 10% heat-inactivated bovine calf serum and 5 × 10⁵ M 2-mercaptoethanol. The cells were then pelleted by centrifugation (500 × g, 10 min, 4°C), the supernatant was decanted, and the cells were resuspended in PBS at a concentration of 2 × 10⁷ cells per ml. MAb 14-4-4s in DMEM was added to the cell suspensions at a final dilution of 1:10 and incubated for 30 min at 4°C with periodic mixing. After incubation, cells were washed twice by centrifugation (500 × g, 10 min, 4°C) and resuspended in DMEM. Cell suspensions were then poured over 100- by 20-mm tissue culture dishes coated with 100 µg of unconjugated goat anti-mouse immunoglobulin in coupling buffer and incubated for 60 min at 4°C. Nonadherent cells were collected by gently rinsing the tissue culture dishes twice with 10 ml of cold DMEM. The cell suspensions from several tissue cultures dishes were aspirated, pooled, and centrifuged at 500 × g for 10 min at 4°C. The supernatant was decanted; the pellet was resuspended in DMEM, poured over another set of immunoglobulin-coated plates, and incubated for 60 min at 4°C. This process was repeated three times. Following the last panning cycle, nonadherent T lymphocytes were washed twice by centrifugation with PBS (500 × g, 10 min, 4°C) and resuspended in PBS. Cell viability was determined by trypan blue (Sigma) exclusion. Cells obtained using this method were shown to contain >95% T lymphocytes by flow cytometry (11, 23).

Isolation of B lymphocytes. Cells remaining on the goat anti-mouse immunoglobulin-coated tissue culture dishes were rinsed twice with 10 ml of cold DMEM to remove any residual nonadherent cells. Five milliliters of cold, nonenzymatic cell dissociation solution was added to the dish, which was incubated at 4°C for 30 min. B lymphocytes were then removed by vigorously tapping and gently scraping the inside of the tissue culture dish with a sterile rubber policeman. Recovered B lymphocytes were pooled and centrifuged at 500 × g for 10 min at 4°C. After centrifugation, the supernatant was decanted and the B lymphocytes were resuspended in cold PBS. B-lymphocyte viability was determined by trypan blue exclusion. Cells obtained by this method were shown to contain 96 to 99% B lymphocytes by flow cytometry.

Analysis of T- and B-lymphocyte preparations by flow cytometry. One-hundred-microliter samples containing 10⁶ lymph node cells were stained before and after panning with MAb 14-4-4s for the presence of B lymphocytes or CD4⁺ T lymphocytes. Cell preparations were stained 1:100 for 15 min at 4°C with a phycoerythrin-conjugated goat anti-hamster immunoglobulin, specific for both heavy and light chains (Boehringer Mannheim Biochemicals). CD4⁺ T lymphocytes were stained with a phycoerythrin-conjugated rat anti-mouse CD4 (L3T4) antibody (1:100; Boehringer Mannheim Biochemicals) for 15 min at 4°C. Samples were then washed twice with PBS by centrifugation, fixed with 1% paraformaldehyde (Sigma), and kept in the dark until analyzed by flow cytometry. Controls used included phycoerythrin-conjugated rat and goat antibodies and unstained T- and B-cell preparations. All samples were analyzed by using a FACScan flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, Calif.). Cells were detected by forward scatter, side scatter, and phycoerythrin fluorescence. Five thousand events were acquired for each sample and were analyzed by histogram profiles of phycoerythrin fluorescence by using LYSYS II software. Gates were drawn using control preparations of unstained samples or cells stained with the isotype control antibodies. The percentage of B lymphocytes present in the cellular suspensions was determined by the percent shift in the phycoerythrin fluorescence of the stained cells.

Production of antibody in vitro. Macrophages (5 × 10⁶) and 2 × 10⁸ B. burgdorferi 297 organisms were cocultured with 5 × 10⁶ T and 5 × 10⁶ B cells from vaccinated or nonvaccinated hamsters in DMEM in sterile flat-bottom multiwell tissue culture plates (Becton Dickinson, Lincoln Park, N.J.) at 37°C with 7.5% CO₂ for 10 days. At days 4, 8, and 10 after cultivation, 0.5-ml samples of the supernatants were collected and centrifuged at 12,000 rpm for 10 min to remove spirochetes and other cellular debris. The supernatants were then stored at 70°C until used. In other experiments, 5 × 10⁶, 2 × 10⁶, 1 × 10⁶, or 5 × 10⁵ T cells from hamsters vaccinated for 7 or 21 days were added to the suspensions of macrophages and B. burgdorferi containing T and B cells from hamsters vaccinated for 14 days. Controls included naive macrophages and B. burgdorferi alone, naive macrophages and B. burgdorferi with naive or immune T cells, naive macrophages and B. burgdorferi with naive or immune B cells, naive or immune B or T cells alone, and naive macrophages with naive or immune T and B cells.

Borreliacidal assay. The frozen supernatants were thawed and serially diluted twofold (1:2 through 1:10,240) in fresh BSK. One-hundred-microliter aliquots of each dilution were transferred to 1.5-ml screw-cap tubes (Sarstedt), and 100 µl of BSK containing 10⁶ B. burgdorferi 297 organisms per ml was added. Subsequently, 10 µl of sterile guinea pig complement (Sigma) was added to each tube. Tubes were vortexed briefly and incubated for 18 to 20 h at 32°C.

9

Western immunoblotting. B. burgdorferi 297 spirochetes were grown in 1 liter of BSK to log phase, pelleted by centrifugation (10,000 × g, 15°C, 10 min), and washed three times with PBS at pH 7.4. The washed pellet was resuspended in 1% formalin and incubated at 32°C for 30 min with periodic mixing and then washed three times by centrifugation (12,000 × g, 10°C, 15 min) and resuspended in PBS. The borrelial protein content was measured by using a bicinchoninic acid assay (Sigma), and the spirochetes were suspended in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer. One hundred twenty micrograms of B. burgdorferi 297 lysate was loaded on a preparative 12% acrylamide gel, and the proteins were resolved by electrophoresis at 18 mA for 5.5 h. The proteins were transferred onto a nitrocellulose membrane for 1 h at 15 V, using a semidry blotting apparatus (Bio-Rad, Hercules, Calif.). The nitrocellulose membrane was incubated for 2 h in 3% milk dissolved in Tris-buffered saline with 0.05% Tween 20 (TBS-T; pH 7.4) to block nonspecific reactivity and then washed 2 times each with TBS-T and double-distilled H₂O. The membrane was then incubated 1 h with a 1:5 dilution of the supernatant (diluted in TBS-T) collected from the suspension of macrophages and B. burgdorferi containing T and B cells from hamsters vaccinated for 14 days. Subsequently, the membrane was washed two times with TBS-T and incubated 1 h with a 1:1,000 dilution of an alkaline phosphatase-labeled goat anti-hamster immunoglobulin G (Kirkegaard & Perry Laboratories, Gaithersburg, Md.) in TBS-T. This was followed by three washes with TBS-T. Antibody binding was detected by the addition of bromochloroindolyl phosphate-nitroblue tetrazolium for 3 min.

Expression and purification of OspA. OspA was purified as previously described (28). Transformed Escherichia coli containing the B. burgdorferi ospA gene was grown in 2× tryptone yeast extract broth containing ampicillin at 37°C for 12 h. Cultures were then diluted with fresh broth and incubated for an additional hour. Isopropyl--D-thiogalactopyranoside was added, and cultures were incubated for 5 h. Subsequently, bacteria were pelleted by centrifugation, resuspended in PBS, and lysed by sonication. Lysed organisms were mixed with Triton X-100, diluted with PBS, and centrifuged again to remove insoluble materials. The supernatant was mixed with a slurry of glutathione-Sepharose beads (Pharmacia) and washed with ice-cold PBS. Fusion proteins were eluted by mixing beads with Tris-HCl containing reduced glutathione and collected after centrifugation. The elution procedure was repeated four times, and the fractions were analyzed for purity by SDS-PAGE and Western immunoblotting.

Preparation of OspA column. Three samples of the OspA fusion protein containing >1.29 mg of protein each were combined and dialyzed overnight in coupling buffer (0.1 M NaHCO₃, 0.5 M NaCl [pH 8.3]). Following dialysis, 0.8 g of CNBr-activated agarose beads was weighed and soaked for 1 h in coupling buffer. Swollen beads were centrifuged (13 min, 1,500 rpm), and the buffer was drawn off. The dialyzed OspA fractions were then added to the activated beads and rocked gently for 2 h at room temperature. After incubation, beads were washed twice with coupling buffer, and a 100 mM solution of ethanolamine (pH 9) was added for 2 h with rocking. The beads were then washed three times with coupling buffer (pH 4), resuspended in a small amount of PBS, and packed into a 5-ml column. Supernatants containing OspA antibodies were placed onto the column and run through approximately 10 times. The loss of OspA antibodies was confirmed by Western immunoblotting, and the treated supernatants were then tested for borreliacidal activity by using the borreliacidal antibody assay.

Statistical analysis. A t test was used to determine differences in the titers of the borreliacidal antibody among supernatants obtained from suspensions of macrophages and B. burgdorferi containing T and B cells from hamsters 14 days after vaccination and those obtained from suspensions of macrophages and B. burgdorferi containing T and B cells from hamsters 7 or 21 days after vaccination and the other controls. The level of significance was set at 0.05 prior to the start of experiments.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Production of borreliacidal antibody in vitro. Suspensions of macrophages and B. burgdorferi were incubated for 4, 8, or 10 days with T and B cells obtained from hamsters 7, 14, or 21 days after vaccination (Fig. 1). Borreliacidal antibody was detected in the supernatants of suspensions of macrophages and B. burgdorferi containing T and B cells obtained from hamsters 14 days after vaccination (Fig. 2). High levels of borreliacidal antibody were detected on days 8 and 10 of in vitro cultivation. By contrast, little (titer of 4) or no borreliacidal antibody was detected in supernatants of suspensions of macrophages and B. burgdorferi containing T and B cells from hamsters 7 or 21 days after vaccination. When these experiments were repeated thrice, borreliacidal activity was detected only when T and B cells were obtained from hamsters 14 days after vaccination.

View larger version (14K): [in this window] [in a new window]   FIG. 1.   Outline of the procedure for production of borreliacidal antibody in vitro by incubating macrophages with B. burgdorferi and T and B cells.

View larger version (53K): [in this window] [in a new window]   FIG. 2.   Production of borreliacidal antibody in suspensions of macrophages and B. burgdorferi incubated for 4, 8, or 10 days with T and B cells obtained from hamsters 7 (), 14 (&atyp0220;), or 21 (&atyp0220;) days after vaccination. The titers of borreliacidal antibody were determined by inoculating dilutions of the cell-free suspensions with B. burgdorferi and complement, incubating for 20 h at 32°C, adding acridine orange, and determining the number of fluorescence-stained B. burgdorferi by flow cytometric analysis. Controls included suspensions containing macrophages and B. burgdorferi, macrophages, B. burgdorferi, and immune cells. When the borreliacidal assays were repeated, results were identical or varied by only a dilution.

Inhibition of the borreliacidal antibody response by T cells. We next determined that T cells obtained from hamsters 7 or 21 days after vaccination could inhibit the in vitro production of borreliacidal antibody (Table 2). When suspensions of macrophages with B. burgdorferi containing T and B cells from hamsters 14 days after vaccination were coincubated with 5 × 10⁶, 2 × 10⁶, 1 × 10⁶, or 5 × 10⁵ T cells from hamsters 7 or 21 days after vaccination, little borreliacidal activity (titer of less than 20) was detected in the supernatants of the suspensions. By contrast, considerable borreliacidal antibody (titer of 1,280) was produced in cultures containing macrophages and B. burgdorferi coincubated with T and B cells from hamsters 14 days after vaccination. When these experiments were repeated thrice, T cells obtained from hamsters 7 or 21 days after vaccination inhibited the production of borreliacidal antibody when cocultured with macrophages and T and B cells obtained from hamsters 14 days after vaccination.

Absorption with recombinant OspA. Cell-free supernatant fluid from suspensions of macrophages and B. burgdorferi containing T and B cells from hamsters 14 days after vaccination were absorbed with recombinant OspA. Little borreliacidal activity was detected in the absorbed preparations. The borreliacidal antibody titer decreased from 1,280 to 4 after absorption with OspA. In addition, the supernatants were tested by Western immunoblotting before and after absorption with OspA. Prior to absorption protein bands of 22, 31, and 34 kDa were detected. After absorption the 31-kDa protein was not detected; however, the 22- and 34-kDa proteins of B. burgdorferi were still observed on the immunoblots.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Considerable research has focused on the development of an effective Lyme borreliosis vaccine. Early studies demonstrated that vaccination with whole B. burgdorferi spirochetes induced protective antibodies in experimental animals (19). These protective antibodies are borreliacidal (8, 28, 36), and the level of borreliacidal antibody is related to the duration of protection (7, 18, 32, 36). Several Osps, especially OspA, of B. burgdorferi have been shown to induce protective borreliacidal antibody in humans and experimental animals (3, 8, 28-30, 34, 35). To date, OspA has received the most intense evaluation as a potential vaccinogen (12-15, 21, 28, 32, 37). However, the anti-OspA borreliacidal antibody response wanes rapidly in OspA-vaccinated humans and hamsters (32). These latter results are discouraging.

In this report, we established that T cells affected the production of borreliacidal antibody. Borreliacidal antibody was detected only in supernatants of cultures containing T and B cells obtained from hamsters 14 days after vaccination. When T cells from hamsters 7 or 21 days after vaccination were coincubated with immune T cells and borreliacidal antibody-producing cells, borreliacidal antibody production was abrogated. These results suggest that immune T cells rapidly lose the ability to promote borreliacidal antibody production, despite the presence of antigen-processing cells and borrelial antigen in the cultures. In addition, T cells obtained 1 week after vaccination suppressed the production of borreliacidal antibody. These findings support our previous study (32) showing that borreliacidal antibody, specifically anti-OspA borreliacidal antibody, develops slowly and wanes rapidly in vaccinated humans and hamsters. We addressed this further by vaccinating C3H/HeJ mice with several whole-cell vaccines (unpublished data). Again, borreliacidal antibody production developed slowly, peaked at weeks 4 to 6 after vaccination, and then decreased rapidly. Our in vitro results showed that the borreliacidal antibody production was directed primarily against OspA because absorption of the cell-free supernatants of 14-day immune T and B cells lost nearly all of their borreliacidal activity. Collectively, these results suggest that T cells play a major role in the ability of OspA to induce and maintain a sustained level of borreliacidal antibody.

Creson et al. (10) reported that borreliacidal activity waned with the elimination of spirochetes from the host by immune clearance or therapy. Our in vitro results extend these findings by demonstrating that borreliacidal antibody production can also wane in the presence of high concentrations of B. burgdorferi spirochetes. The mechanism responsible for the delay in production and waning of anti-OspA borreliacidal antibody involves T cells. T cells from hamsters 7 or 21 days after vaccination prevented anti-OspA borreliacidal antibody production when cocultured with T and B cells obtained from hamsters 14 days after vaccination. This effect was also detected when either 7- or 21-day T cells were added in reduced concentrations to the borreliacidal antibody-producing 14-day immune T and B cells. In support, 2 × 10⁸ spirochetes were present throughout the duration of the in vitro cultivation of suspensions of T and B cells obtained from hamsters 14 days after vaccination and cocultured with T cells obtained from hamsters 7 or 21 days after vaccination. These findings demonstrate that T cells or their products play a major role in influencing the induction and more importantly the decline of protective borreliacidal antibody, despite the presence of B. burgdorferi antigens. Vaccine strategies, therefore, must place more emphasis on defining the role that immune cells play in these responses.

Previously, we showed that B. burgdorferi-specific T lymphocytes were also responsible for the induction of severe destructive Lyme arthritis (22, 23). When naive hamsters were infused with T lymphocytes from vaccinated hamsters, they developed severe destructive arthritis after challenge with B. burgdorferi unless high levels of borreliacidal antibody were present at the time of infection. The ability of T cells or their products to decrease borreliacidal production will likely make frequent boosters or the addition of a safe adjuvant necessary to prolong the high levels of borreliacidal antibody. However, repeated vaccinations and use of an adjuvant may increase the potential for side effects (1, 16, 20, 39), such as severe destructive arthritis (21, 22). Thus, additional experiments are needed to determine the immunologic mediator(s) responsible for maintaining sustained high levels of borreliacidal antibody.

Our findings are significant for an additional reason. Although anti-OspA borreliacidal antibodies were readily detected in human volunteers after primary and booster vaccination, the levels of borreliacidal activity varied widely and decreased rapidly (32). Only one vaccinee had detectable borreliacidal activity 6 months after vaccination. Keller et al. (21) suggested that an anamnestic response would provide protection against infection even in the absence of circulating antibodies. Our in vitro results suggest that the circulating antibody is the result of a limited production of borreliacidal antibody by immune cells obtained 14 days after vaccination. Once borreliacidal antibody production is down-regulated (21 days after vaccination), immune cells failed to produce borreliacidal antibody, even in the presence of antigen-processing cells and borrelial antigen. This finding suggests that immune T cells become less responsive or tolerant to epitopes of OspA responsible for the induction of borreliacidal antibody. In support, we showed previously (32) that production of anti-OspA borreliacidal antibody did not correlate with production of total OspA antibody after vaccination of humans or hamsters with OspA. Furthermore, Foley et al. (15), Straubinger et al. (48), and Schutzer et al. (37) showed that infection with B. burgdorferi occurred in OspA-vaccinated rabbits, dogs, and a human, respectively. Again, OspA vaccination may be of limited value considering the heterogeneity of OspA (28) and the inability of the protective epitope of OspA to induce high and sustained levels of borreliacidal antibody.

Our results also show that other specific anti-B. burgdorferi antibodies, other than anti-OspA borreliacidal antibody, can be detected in cultures of macrophages and B. burgdorferi containing T and B cells from hamsters 14 days after vaccination. When cell-free supernatants from these cultures were absorbed with recombinant OspA, little borreliacidal activity was detected. However, anti-B. burgdorferi antibodies against the 22- and 34-kDa proteins were detected in the absorbed samples by Western immunoblotting. Although the 22- and 34-kDa proteins have been shown to induce protective antibodies (14, 17, 33, 34), our in vitro results indicate that the protective or borreliacidal epitopes of these Osps were not processed in vitro or in vivo by macrophages for presentation to immune T and B cells. One would expect that in vitro exposure of B. burgdorferi to antigen-presenting cells (macrophages) and immune T and B cells obtained from hamsters 14 days after vaccination would augment or induce a protective borreliacidal response to the 22- and 34-kDa proteins. Therefore, our results indicate that the putative protective borreliacidal epitopes, including those of OspA, are difficult for the host's immune system to recognize, despite vaccination of hamsters with B. burgdorferi contained in an adjuvant and subsequent exposure in vitro of these immune T cells to antigen-processing cells and high concentrations of spirochetes.

In conclusion, we showed that anti-OspA borreliacidal antibody was produced in vitro only when T and B cells were obtained from hamsters 14 days after vaccination. Production of the anti-OspA borreliacidal antibody was prevented by the addition of T cells obtained from hamsters 7 or 21 days after vaccination. Further studies are needed to delineate the mechanism(s) of induction and waning of borreliacidal activity. These studies will improve the immunogenicity of OspA and other Osps and aid in developing an efficacious and safe vaccine against infection with B. burgdorferi.