INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Lyme disease, which is caused by infection with the spirochete Borrelia burgdorferi, is a vector-borne illness acquired through the bite of some Ixodes ticks (2, 34, 36, 37). The duration of tick attachment to the vertebrate host determines whether spirochetes are successfully transmitted, with the infection risk being greatest after 48 h (33, 35). After tick attachment, spirochetes previously localized to the midgut multiply exponentially and migrate to the salivary gland, where they exit the tick through salivary secretions (8, 25, 27). Spirochete growth is accompanied by marked phenotypic changes in outer surface protein (Osp) expression. In particular, spirochetes modulate expression of OspA (8, 31), a surface lipoprotein that has been reported to mediate spirochete binding to the tick midgut (24). Spirochetes in the midguts of unfed ticks uniformly express OspA, and replicating spirochetes in the midguts of feeding ticks are a mixed OspA⁺ and OspA population, whereas the majority of those in the salivary glands of feeding ticks no longer express this protein (7, 23). After deposition within the mammal, lipoproteins can activate innate immune cells through pattern recognition receptors and induce inflammation (3, 15). Thus, spirochete modulation of surface lipoprotein expression may serve several purposes, including facilitating migration within the tick and transmission to the vertebrate host as well as evasion of the host innate immune response.

One of the first immune components encountered by spirochetes within the midguts of ticks feeding on naive hosts are natural antibodies. Natural antibodies are germ line-encoded molecules produced by a distinct population of peritoneal B cells bearing the cell surface marker CD5 and are present in the sera and interstitial fluids of healthy animals (14, 16, 17). The majority of natural antibodies are immunoglobulin M (IgM) isotype; are polyreactive, with various affinities for multiple antigens, including pathogens and toxins (5, 22); and are present even in the absence of exposure to environmental pathogens (6, 17). Although their contribution to immune defense has only recently been appreciated, at concentrations present in serum, natural antibodies are able to kill bacteria in vitro (22) and help clear lipopolysaccharides in vivo (21, 26). They also facilitate uptake, processing, and presentation of antigen by B cells (5, 38) and may help localize pathogens and their antigens to lymphoid organs (14, 16, 22). B-cell-deficient mice, which lack natural antibodies, have an increased pathogen burden in nonlymphoid organs when infected with viruses or intracellular bacteria compared to wild-type mice (22). Taken together, these findings suggest a role for natural antibodies in limiting the initial pathogen burden prior to the development of adaptive immune responses (22).

Indirect evidence suggests that natural antibodies could contribute to the host defense against B. burgdorferi infection. It has been shown that IgM specifically interacts with OspA present in in vitro spirochete cultures (11, 41). Sera from several species of nonimmune animals, which contain natural antibodies, can kill spirochetes in vitro, and the killing is complement dependent (18). In this study, we postulated that ingestion of natural antibodies during tick feeding could influence survival of spirochetes within an infected tick, before their deposition within the mammal. Because OspA is a dominant antigen on cultured spirochetes as well as those in the tick midgut where they first encounter natural antibodies, those spirochetes within the tick midgut may be particularly susceptible to the effects of natural antibodies. To investigate these issues, we compared the prevalence of OspA-expressing spirochetes in the salivary glands as well as the total spirochete burden within ticks that fed on B-cell-deficient and normal mice. Our results implicate natural antibodies of the IgM class in limiting survival of spirochetes in feeding ticks.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Mice. Six-week-old C57BL/6J (B6) and B-cell-deficient B6.129S2-Igh^/ (B6.Igh^/) mice were purchased from Jackson Laboratories. Mice were housed in filter frame cages and administered food and water ad libitum according to Yale University animal care and use guidelines. Mice were euthanized by carbon dioxide asphyxiation.

Tick infection. Nymphal ticks infected with the pathogenic B. burgdorferi N40 strain (20) were derived from a specific-pathogen-free laboratory-reared colony of Ixodes scapularis ticks maintained at the Department of Epidemiology and Public Health at Yale University. Larval ticks were infected by feeding on mice inoculated intradermally 1 month previously with 10⁴ cultured N40 organisms. After feeding to repletion and detaching naturally, engorged larvae were retrieved and housed in 22°C humidified environmental chambers until they molted into nymphs. The rate of infection of nymphs was 80 to 90% as assessed by direct immunofluorescence of spirochetes in midgut extracts of representative ticks.

Passive immunization of mice. Mouse serum IgG and mouse monoclonal IgM (mIgM) (kappa clone TEPC 183, which recognizes an unidentified mouse tumor antigen) were purchased from Sigma Chemical Co. (St. Louis, Mo.). Polyclonal IgM (pIgM) was generated by depleting IgG from B6 mouse serum using the monoclonal antibody (MAb) Trap II kit according to the instructions of the manufacturer (Amersham-Pharmacia Biotech, Piscataway, N.J.). Briefly, whole normal mouse serum (NMS) diluted 1:3 in phosphate-buffered saline (PBS) was passed over a HiTrap protein G column twice, and the eluant was collected. The IgG-depleted eluant was buffer exchanged into PBS and concentrated using a Centriprep p-30 column (Amicon, Bedford, Mass.). The IgM and IgG concentrations were determined by Ig isotype-specific enzyme-linked immunosorbent assay using biotinylated secondary antibodies and the Vectastain Elite and ABTS [2,2'-azinobis(3-ethylbenzthiazolinesulfonic acid)] kits (Vector Laboratories, Burlingame, Calif.) according to the manufacturer's protocols. The NMS used in these experiments contained 1.4 mg of IgG per ml and 1.1 mg of IgM per ml, and pIgM contained 0.2 mg of IgM per ml and 0.003 mg of IgG per ml.

Salivary gland microscopy. The salivary glands from individual nymphs were harvested under a dissecting microscope and fixed by immersion in 50 µl of 4% formaldehyde in PBS on silylated glass slides (PGC Scientific, Gaithersburg, Md.) for 1 h at room temperature. Tissues were rinsed twice with PBS, permeabilized by incubation in 0.5% Triton X-100 in PBS for 15 min at room temperature, and then blocked for 1 h at room temperature or overnight at 4°C in PBS containing 10% normal goat serum. For immunofluorescent staining, fixed and permeabilized salivary glands were incubated for 1 h with B-cell hybridoma supernatant containing anti-OspA MAb VIIIC3.78 (32) diluted 1:75 in PBS-10% normal goat serum. The organs were washed three times with PBS and then incubated for 1 h with fluorescein isothiocyanate (FITC)-conjugated goat anti-B. burgdorferi IgG (1:50 dilution) (KPL, Gaithersburg, Md.) and Texas red-conjugated anti-mouse IgG (1:100 dilution) (Tago, Burlingame, Calif.). After extensive washing, stained organs were air dried and mounted onto glass slides using Gel/Mount (Biomeda Corp., Foster City, Calif.). Samples were imaged using an LSM 510 scanning laser confocal microscope equipped with an argon-krypton laser (Carl Zeiss Inc., Thornwood, N.Y.). The salivary glands were visualized using the confocal microscope 63× objective, and the section depth was 0.6 µm.

Competitive PCR of B. burgdorferi DNA. Whole ticks were ground in Eppendorf tubes using disposable plastic pestles, and the DNA was extracted using the Isoquick DNA isolation kit (Orca Research Inc., Bothwell, Wash.). The DNA samples were resuspended in 50 µl of water. The relative amount of B. burgdorferi DNA in each sample was determined using a competitive PCR targeting a B. burgdorferi-specific fragment of the chromosomal glyceraldehyde-3-phosphate dehydrogenase gene (termed Bb-GAPDH). Based on GenBank data, the B. burgdorferi GAPDH gene has homology only to that of a related spirochete, Borrelia hermsii, the agent of relapsing fever (13). The Bb-GAPDH primers used were as follows: 5' primer ^5'AGTCAAGAGATGGTGCCATTGTTGTGG^3' (bp 319 to 345) and 3' primer ^5'GTACCATTCAACTTACCCTTAAGTTCGG^3' (bp 817 to 844). The competitor gene target was generated by cloning the full-length GAPDH gene into plasmid pGEX-5x-2 (Amersham-Pharmacia Biotech) and excising bp 402 to 471. Amplification of the competitor gene using the Bb-GAPDH primers results in a 456-bp product, which is 69 bp smaller than that amplified from the native gene (525 bp). One microliter of tick DNA was amplified in a 50-µl total volume containing 25 µM primers, 1.6 mM deoxynucleoside triphosphates (Boehringer Mannheim, Indianapolis, Ind.), 2.75 mM MgCl₂, and 0.5 µl of Taq polymerase (Qiagen, Valencia, Calif.) using a Robocycler (Stratagene, La Jolla, Calif.) with the following settings: denaturation at 95°C for 90 s, annealing at 56°C for 60 s, and extension at 73°C for 75 s for a total of 40 cycles. Titrations of known quantities of the competitor plasmid were added to each reaction mixture, and the relative amount of spirochete DNA in each sample was determined from the quantity of competitor added to each PCR that resulted in equivalent amplification of competitor and Bb-GAPDH target gene sequences. This competitive assay was validated using DNA from known titrating quantities of cultured B. burgdorferi (data not shown). Bb-GAPDH PCR products were normalized to the amount of tick actin amplified from each sample. A 400-bp fragment of actin was amplified from 1 and 0.25 µl of sample DNA using the following arthropod actin-specific primers: 5' primer ^5'GCCGATGGTGATCACCTGTCCG^3' and 3' primer ^5'GATGACCCAGATCATGTTCGAGCC^3'. Reaction conditions were similar to those used for Bb-GAPDH, and the thermocycler settings were as follows: denaturation at 95°C for 90 s, annealing at 59°C for 60 s, and extension at 73°C for 75 s for a total of 36 cycles.

In vitro borrelicidal assay. Low-passage N40 spirochetes were grown to logarithmic phase in Barbour-Stoenner-Kelley (BSK) II medium (1) and then washed and resuspended at a final concentration of 10⁷ spirochetes per ml in PBS supplemented with 5.4 mM glucose, 50% heat-inactivated rabbit serum, and A 1956, an antibiotic mixture designed for Borrelia culture (Sigma). Spirochetes remain viable but do not divide in this solution. Spirochetes (2 × 10⁶/200 µl) were aliquoted in triplicate into screw-top Eppendorf tubes containing the indicated dilution of MAb or serum (see Table 1) and 10 µl of mouse complement (1 50% hemolytic complement unit/ml; Sigma). The final amounts of Igs in the samples were as follows: (i) for MAb VIIIC3.78, 0.16 µg of IgG; (ii) for B6 14-day immune serum, 12.8 µg of IgG and 8.8 µg of IgM; and (iii) for B6 naive serum, 11.6 µg of IgG and 8.8 µg of IgM. After 24 h of incubation at 33°C, samples were examined by dark-field microscopy for viable spirochetes in a double-blind fashion. Spirochetes were considered dead when complete loss of motility and refractivity was observed. Spirochetes were enumerated in 10 visual fields, and the percent viability was calculated as the ratio of live spirochetes (mean of 10 fields) in treated samples to spirochetes in the untreated control samples (mean of 10 fields). Serum from 14-day B. burgdorferi-infected B6 mice and the borrelicidal MAb VIIIC3.78 were used as positive controls for killing. In addition, from the third experiment, a 50-µl aliquot of spirochetes from each tube was inoculated into 0.5 ml of BSK II medium and incubated at 33°C for 48 h. The viable spirochetes in the cultures were enumerated, and percent viability was calculated as described above.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Natural IgM antibody can kill spirochetes in vitro. Previous studies have shown that IgM can bind to cultured spirochete Osps (11) and that serum from nonimmune mammals can kill spirochetes in vitro (18). To determine whether natural antibodies could account for the borrelicidal activity, we compared the viabilities of spirochetes cultured in the presence or absence of natural antibody using serum from control B6 mice or serum from B-cell-deficient mice (B6.Igh^/), which do not produce antibody (Table 1). In this assay, spirochetes were readily killed by the borrelicidal anti-OspA MAb and serum from 14-day-infected B6 mice (Table 1). In addition, naive B6 serum was borrelicidal, albeit with reduced efficiency compared to immune sera. In contrast, serum from B6.Igh^/ mice had no effect on spirochete survival. Addition of pIgM to B6.Igh^/ serum restored the killing capacity of this serum, but addition of IgG had only a minimal effect. Based on the number of remaining viable spirochetes, the killing capacities of only the OspA MAb, immune serum, naive serum, and B6 Igh^/ serum supplemented with pIgM were found to be statistically different from that of the PBS negative control (P < 0.01 using Dunnett's multiple-comparison test). Spirochetes cultured with pIgM without B6.Igh^/ serum and complement were not killed, indicating that pIgM did not contain significant levels of complement or other non-Ig serum components with potential borrelicidal activity (data not shown). Aliquots of spirochetes from the third experiment were cultured in BSK II medium for 2 days. Enumeration of the spirochetes at this time point confirmed the results obtained from the borrelicidal assay, with similar relative differences in the percentages of viable spirochetes among the samples.

Spirochetes that migrate to the salivary glands of ticks that feed on B6.Igh^/ mice express OspA. Previous studies have shown that the majority of spirochetes within the salivary glands of feeding ticks no longer express OspA (7, 9). Our findings that serum containing pIgM can kill OspA-expressing spirochetes in vitro suggested that natural antibodies could influence survival of OspA-expressing spirochetes within feeding ticks. Natural antibodies that bind to OspA-expressing spirochetes in the midgut may promote their elimination in the gut or hemolymph, preventing them from reaching the salivary glands. We therefore sought to determine whether natural antibody influences the percentage of OspA-expressing spirochetes in the salivary glands of feeding ticks. Forty-eight hours after feeding on B6.Igh^/ or control mice, ticks were removed and the spirochetes within salivary glands were visualized by confocal immunofluorescence microscopy. Using two-color staining, this technology permits simultaneous identification of spirochetes with anti-B. burgdorferi antibodies and assessment of their OspA expression within the thick tissue of the salivary gland. As previously reported, OspA⁺ spirochetes were only rarely detected in salivary glands of ticks that fed on control B6 mice (Fig. 1, WT). In contrast, examination of multiple salivary glands from ticks that fed on B6.Igh^/ mice revealed that almost all spirochetes could be visualized using the OspA MAb (Fig. 1, PBS).

View larger version (65K): [in this window] [in a new window]   FIG. 1.   Confocal imaging of OspA expression on spirochetes within tick salivary glands at 48 h of feeding. Each frame shows lobes of a representative salivary gland from each mouse group. B6.Igh/ mice were pretreated as described in Materials and Methods with PBS, IgG, mIgM, pIgM, or NMS. Control B6 mice (wild type [WT]) were untreated before tick placement. OspA was detected on spirochetes within salivary glands by indirect immunofluorescence using the anti-OspA MAb VIIIC3.78 and Texas red-conjugated anti-mouse IgG. Total spirochetes were detected using FITC-conjugated polyclonal goat anti-B. burgdorferi IgG. For each mouse group, the three panels show confocal images of the same field using the Texas red, FITC, and merged channels. The results shown here are representative of those from five separate experiments. Magnification, ×63.

Natural IgM but not IgG decreases the prevalence of OspA⁺ spirochetes in salivary glands of ticks feeding on B6.Igh^/ mice. We next passively immunized B6.Igh^/ mice with NMS, pIgM, mIgM, IgG, or PBS prior to tick placement in order to assess which component of natural antibody, if any, influenced the presence of OspA⁺ spirochetes in the salivary gland. At 48 h of feeding, ticks that fed on IgG- or mIgM-immunized B6.Igh^/ mice had predominantly OspA⁺ spirochetes within their salivary glands (Fig. 1). However, OspA⁺ spirochetes were rarely seen in the salivary glands of ticks feeding on NMS-immunized or pIgM-immunized B6.Igh^/ mice, similar to what was observed in ticks feeding on B6 control mice (Fig. 1). The number of spirochetes per gland was relatively small at the 48-h time point, even when ticks fed upon B6.Igh^/ mice. In order to determine the percentages of OspA-expressing spirochetes in the salivary glands, we analyzed and counted the total number of spirochetes in four to six glands from each experimental group by confocal microscopy (Table 2). The greatest percentages of spirochetes that express OspA in salivary glands were found in ticks that fed on B6.Igh^/ mice treated with PBS alone, mIgM, or IgG (Table 2). In contrast, OspA was found on fewer than 10% of the spirochetes in salivary glands from ticks that fed on B6.Igh^/ mice treated with NMS or pIgM and in salivary glands from ticks that fed on normal B6 mice (Table 2).

OspA is not detected on spirochetes in tick salivary glands after 72 h of feeding, independent of antibody. The mechanism by which spirochetes diminish OspA expression within the tick has not been clearly defined. However, there is no evidence showing that an individual spirochete can regulate Osp expression at the molecular level because of immune pressure. Thus, we expect that OspA expression can still be modulated by spirochetes regardless of the host on which the tick obtains its blood meal. Indeed, OspA-expressing spirochetes were rarely observed in salivary glands of ticks that were allowed to feed for 72 h, irrespective of the mouse host, with the PBS-immunized B6.Igh^/ group having the largest percentage remaining (Table 2).

Natural antibody reduces the total number of spirochetes within feeding ticks. Cultured B. burgdorferi strain N40 spirochetes express high levels of OspA (19) and therefore are similar in phenotype to tick midgut spirochetes, which also express OspA (9). We have shown that OspA-expressing spirochetes are susceptible to the borrelicidal activity of natural IgM in vitro and that there is a marked decrease in the number of OspA-expressing spirochetes in salivary glands of ticks that fed on B6 mice as well as in those that fed on B6.Igh^/ mice treated with NMS or pIgM. Taken together, these data suggest that IgM natural antibody may reduce the number of spirochetes replicating in the midguts of feeding ticks, prior to their migration to the salivary glands. We next used a competitive PCR to assess the relative number of spirochetes in ticks that fed on B6.Igh^/ or control mice. Whole ticks were used because spirochetes may be lost during dissection of individual organs. A B. burgdorferi-specific fragment of the GAPDH gene was selected for amplification because this target is present in a single copy on the B. burgdorferi chromosome. Ticks were removed 72 h after placement, and scutal indices were measured to ensure comparable durations of feeding (12, 35). Compared to ticks that fed on B6 mice, those that ingested a blood meal from B6.Igh^/ mice harbored on average fivefold more spirochete DNA (Table 3). Passive transfer of IgG to B6.Igh^/ mice prior to tick placement did not reduce the pathogen load in feeding ticks. In contrast, ticks that fed on B6.Igh^/ mice passively immunized with NMS or pIgM had levels of B. burgdorferi DNA equivalent to those of ticks retrieved from control B6 mice (Table 3).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

These studies provide evidence that natural antibody exerts an effect on spirochete survival within feeding ticks. Using a competitive PCR for spirochete DNA, we have shown that ticks feeding on B6.Igh^/ mice harbor more spirochetes than those feeding on normal B6 mice. While the absolute numbers of spirochetes in unfed ticks are presumed to vary, we found a reproducible increase in the pathogen burden among engorged ticks removed from B6.Igh^/ mice compared to those feeding on control mice. NMS can kill spirochetes in vitro, and we show that natural antibody of the IgM class is required for the borrelicidal activity both in vitro and in feeding ticks. Although it was not analyzed in this study, we believe that this effect of natural antibodies is preferentially exerted on spirochetes within the midgut, as there is no evidence that host IgM, unlike host IgG, can penetrate the hemolymph and enter the salivary gland (39).

OspA is a dominant surface-exposed lipoprotein on cultured N40 spirochetes (19) as well as those within the midguts of unfed ticks (9). Previous studies have shown that OspA⁺ spirochetes are rarely found within salivary glands of ticks feeding on normal nonimmune hosts (7, 9). We show here that the majority of spirochetes within the salivary glands of ticks feeding on B6.Igh^/ mice continue to express OspA at 48 h after tick attachment. In one experiment, we observed OspA⁺ spirochetes at as late as 64 h of tick feeding (data not shown). In fact, we found that in the absence of natural antibody almost all salivary gland spirochetes (90%) expressed OspA when ticks fed for 48 h, compared to only 5% when ticks fed on normal mice. Other studies confirm our observation that OspA⁺ spirochetes can occasionally be found within the salivary glands of feeding ticks removed from normal B6 mice, albeit at a much lower frequency (23).

The absence of natural antibody, which we show can kill OspA-expressing spirochetes in vitro, may permit more OspA⁺ spirochetes to survive and reach the salivary glands. However, the majority of spirochetes after 72 h of tick feeding no longer express OspA, regardless of the presence of natural antibody in the host. For this reason, we do not believe that the immune pressure exerted by natural antibody directly influences expression of OspA or other surface-exposed Osps at the molecular level. Our studies did not address the molecular mechanisms by which the expression of OspA is controlled.

Our studies utilized MAb VIIIC3.78 to detect OspA on spirochetes. It is possible that natural antibody present in serum could mask the MAb binding site, obscuring the detection of OspA expression on salivary gland spirochetes. However, this would not seem to be a concern, as spirochetes are efficiently killed within ticks allowed to feed upon mice passively immunized with MAb VIIIC3.78 (10), indicating that this MAb can still access or compete for its binding site on OspA in the presence of natural antibodies.

OspA is reported to mediate binding of spirochetes to the tick midgut (24), and presumably that binding must be interrupted for spirochetes to leave the midgut. Our data quantifying OspA⁺ spirochetes in salivary glands show that complete downregulation of OspA is not necessary for spirochetes to leave the midgut and migrate to the salivary glands. In addition, OspA complexed with IgM has been found in patients with early Lyme disease, indicating that some spirochetes continue to express this antigen after transmission to the vertebrate host (4, 28-30). It may be that the tick midgut loses expression of the OspA ligand(s) during tick feeding, resulting in diminished spirochete binding independent of OspA expression.

pIgM, but not IgG, delivered to B6.Igh^/ mice prior to tick placement diminishes the percentage of OspA-expressing spirochetes within tick salivary glands to levels in ticks feeding on normal hosts. The ability of antibody to fix complement may explain the differential effects of natural IgM and IgG in these assays. In addition, tick saliva contains IgG binding proteins that can sequester IgG, possibly to prevent damage to the tick, and spirochetes may indirectly benefit from this tick defense mechanism (40). IgM may exert selective pressure on spirochetes such that those with reduced levels of surface-exposed Osps, including OspA, have much higher survival rates. This may explain the apparent reduced infectivity of OspA⁺ spirochetes deposited within host tissues during the early stages of tick feeding (23), as those spirochetes that escaped midgut elimination will next encounter IgM in the mammalian skin. Binding of IgM to OspA may also expedite shedding of the protein from the surface of the spirochetes. Our study did not examine OspA expression at the molecular level, and OspA appears to be normally downregulated in the late stages of tick feeding.

In summary, this study shows that natural antibody of the IgM class has borrelicidal activity and can reduce spirochete numbers within feeding I. scapularis nymphs. Natural antibodies may therefore function as a first defense against B. burgdorferi and other vector-borne pathogens that have exposure to these molecules in the tick midgut prior to deposition within the vertebrate host.