Arthropod- and Host-Specific Borrelia burgdorferi bbk32 Expression and the Inhibition of Spirochete Transmission

Spirochetes

A clonal isolate of B. burgdorferi N40 that is infectious and pathogenic in mice was used throughout (36). Spirochetes were cultivated in BSK II medium at 33°C (37).

Animals

Three-week-old female C3H/HeNCr (C3H) and C.B-17-scid mice were obtained from the Frederick Cancer Research Center (Frederick, MD) and housed in filter-frame cages. Mice were given food and water ad libitum and euthanized with CO₂.

Two hundred- to 500-g female Hartley guinea pigs were housed in individual hanging cages at Yale University animal facilities.

Ticks

Mated adult female I. scapularis were collected in the field. The egg mass was then laid in the laboratory. Hatched larvae were fed on uninfected C3H mice to produce pathogen-free nymphs. Alternatively, to generate B. burgdorferi-infected nymphal ticks, larvae were allowed to engorge to repletion upon B. burgdorferi N40-infected mice and then molt to the nymphal stage. B. burgdorferi infection rates within the nymphs were >90%. All tick rearing was performed in an incubator at 26°C with 85% relative humidity and a 12-h:12-h light:dark photo period regimen.

Infection and immunization

For infection, 4- to 6-wk-old C3H mice were intradermally challenged with 10⁴ B. burgdorferi N40 and sacrificed at 14, 30, and 60 days. In an experiment, groups of 10–20 mice were inoculated with B. burgdorferi and 4–5 mice were sacrificed at each time point. All studies were repeated, and in each experiment, tissue samples from each mouse were collected and stored individually.

For the postinfection clearance studies, mice were first infected with 10⁴ spirochetes by intradermal syringe inoculation and administered 200 μl of rabbit BBK32 and BBK50 antisera on days 1, 2, 4, or 8 (11). Control mice were administered glutathione transferase (GT) antisera in an identical manner (11). The BBK32 and BBK50 antisera had been previously produced (11). BBK32 antisera were prepared by immunizing animals with the p35 portion of the BBK32 protein and have been shown to be immunogenic in the murine model (11). The full-length BBK32 protein from B. burgdorferi N40 has also been produced in our laboratory, but is not highly antigenic in its current form (data not shown) and was therefore not used for these studies. Mice were killed at 14 days. At necropsy, blood, spleen, urinary bladder, and skin specimens were collected aseptically, cultured in BSK II medium, and examined by dark-field microscopy for spirochetes. The tibiotarsal joints and the hearts were Formalin-fixed, paraffin-embedded, sectioned, and examined microscopically for inflammation (36). Arthritis was graded by blinded analysis on a scale from 0, no inflammation, to 3, severe disease (36).

For the larval acquisition studies, groups of four to five C.B.17-scid mice were infected with 10⁴ B. burgdorferi by intradermal challenge. At 20 days, mice were passively immunized with 200 μl of BBK32 and BBK50 antisera or a control GT antiserum. Twenty-four hours after passive immunization, mice were anesthetized with ketamine/xylazine and 50 larvae were placed on each mouse. The larvae fed to repletion and were examined for spirochetes. Individual larvae were tested by immunofluorescence (see below), using B. burgdorferi-antisera, for the presence of spirochetes.

For the tick-borne transmission studies, groups of four to five mice were first administered 200 μl of BBK32 and BBK50 antisera. Twenty-four hours later, 4 or 20 I. scapularis nymphs, infected with B. burgdorferi N40, were allowed to feed to repletion on each of the mice. Engorged ticks usually fell from the animals within 4 days and were collected in water baths below the cages. Two weeks later, all mice were sacrificed and examined for infection and disease, as previously stated.

In the guinea pig infection studies, a 5- to 10-cm section of the animals’ posterior mid thorax was shaved and then three to four adult ticks were applied per guinea pig and allowed to attach. The adult ticks were obtained from groups of nymphal ticks that had fed upon mice immunized with BBK32, BBK50, or GT antisera and then allowed to undergo the molting process. Guinea pigs were sacrificed 2 wk after tick challenge. Sera were collected and tested for Abs to B. burgdorferi by immunoblot (38).

PCR

PCR amplification of B. burgdorferi DNA was performed for 30 cycles with denaturing, annealing, and extension temperatures of 94°C for 1 min, 55°C for 1 min, and 72°C for 2 min. 5′ and 3′ primers were: 5′-CAATCAGGTAACGGC-3′ and 5′-TCTATTTTGGAAAGC-3′ for flaB, 5′-GC GGTGAAAGTGGTGAATTGG-3′ and 5′-CTTTAGCAAAGTTGTCAAGGCGTG-3′ for bbk32, and 5′-GACAACAACAGAAGTGGTCGCAC-3′ and 5′-TAGCATCGGAA TGAGCTGTACC-3′ for bbk50. flaB, which has one copy on the B. burgdorferi chromosome, was used to quantify the organism load (18).

For RT-PCR, total RNA was isolated from the spleens, skin, tibiotarsal joints, bladders, and hearts of B. burgdorferi-infected mice by acid guanidinium thiocyanate phenol-chloroform extraction (16). RNA was treated with RNase-free DNase (Promega, Madison WI) for 3 h at 37°C. A total of 2 μg of RNA was then reverse transcribed using Moloney murine leukemia virus reverse transcriptase (Stratagene, La Jolla, CA). PCR was then conducted for 30 cycles.

For quantitative PCR, a competition assay was performed based on established techniques (12, 39). Competitor constructs were created in which a 540-bp fragment of unrelated DNA was flanked by the flaB, bbk32, or bbk50 primers. The unrelated DNA was a fragment of the gene encoding the 44-kDa Ag of the agent of human granulocytic ehrlichiosis, amplified using the Ehrlichia primers 5′-TTTGCAGGACGAAGAAGAATGG-3′ and 5′-CCAACGCAAGCATAAGGAACAAC-3′, which have previously been used in constructs for competitive PCR (40). Varying concentrations of competitor were added to the RT-PCR reactions, and the concentration at which the amplified competitor product and the gene of interest were similar in intensity was used to calculate the quantity of mRNA or DNA.

Immunofluorescence of B. burgdorferi within ticks

Groups of B. burgdorferi N40-infected ticks, including unfed ticks and ticks that fed to repletion on mice, were checked for spirochetes, and the presence of specific B. burgdorferi Ags, within the vector by indirect immunofluorescence (IFA). Each tick was dissected to separate the internal organs from the cuticle and lightly homogenized in 100 μl of PBS. Then triplicate 10-μl aliquots were spotted on silylated glass slides. The slides were air dried and fixed with 4% paraformaldehyde. The specimens were incubated in rabbit BBK32, BBK50, OspC, or OspA antisera (1:100 dilution) for 1 h, washed, followed by incubation in anti-rabbit IgG coupled with FITC at a 1:500 dilution for another hour, and viewed under a Zeiss Axioskop fluorescence microscope (Oberkochen, Germany) (11, 27, 41). B. burgdorferi antisera or FlaB antisera, which recognizes B. burgdorferi within both unfed and engorged ticks, was used as positive controls (4, 40). GT antisera was used as the negative control. Ten high-power fields (×400) were examined, and the total number of spirochetes was calculated.

B. burgdorferi bbk32 and bbk50 expression during murine infection

bbk32 and bbk50 have been shown to be induced during murine infection and in several patients with erythema migrans or Lyme arthritis (11, 20). We have now quantitatively examined the temporal and tissue-specific expression of B. burgdorferi bbk32 and bbk50 during experimental murine Lyme borreliosis. Following intradermal challenge, spirochetes are deposited in the skin and then begin disseminating (4–10 days) to the joints, heart, spleen, and bladder (36, 42). Arthritis and carditis then become prominent (2–4 wk) and regress over 1–2 mo (36, 42). At later intervals, mice remain infected but disease is no longer apparent (36, 42). bbk32 and bbk50 mRNA levels were therefore first measured at the cutaneous site of inoculation from days 1 to 14 (Fig. 1⇓) and then at different disseminated sites on days 14, 30, and 60, intervals that reflect the various stages of disease (Table I⇓). flaB mRNA, which is constitutively expressed, was used for comparison (12). Representative examples of the flaB, bbk32, and bbk50 quantitative RT-PCR are shown in Fig. 2⇓.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

bbk32, bbk50, and flab mRNA levels at the initial cutaneous site following challenge with 10⁴ B. burgdorferi via intradermal syringe inoculation. mRNA levels were measured by quantitative RT-PCR and expressed as femtograms per skin specimen. Each skin sample weighed 20 mg.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Representative competitive RT-PCR assays to quantify flaB (A), bbk50 (B), and bbk32 (C) mRNA levels during murine infection with B. burgdorferi. In all panels, the upper band represents the competitor construct, which was added in increasing concentrations from lanes 1 to 6. Amount of competitor: lane 1, 3.2 × 10⁻¹⁰ μg; lane 2, 1.1 × 10⁻⁹ μg; lane 3, 8 × 10⁻⁹ μg; lane 4, 4 × 10⁻⁸ μg; lane 5, 2 × 10⁻⁷ μg; and lane 6, 2 × 10⁻⁶ μg. The lower band represents the amplified B. burgdorferi product.

At the local skin site, following the intradermal inoculation of 10⁴ spirochetes, flaB mRNA was detected (3.8 ± 1.3 femtograms/tissue specimen) on day 1 (Fig. 1⇑). At this time point, quantitative DNA-PCR revealed that 2 × 10⁴ ± 0.5 × 10⁴ spirochetes were present at the site of inoculation, representing the original inocula and perhaps some initial spirochete growth. By day 6, flaB mRNA levels increased 10-fold (37 ± 10 femtograms/tissue specimen) from the initial values before declining on day 14. Quantitative DNA-PCR demonstrated that the spirochete number increased up to 3 × 10⁵ ± 0.6 × 10⁵ on day 6 before declining, reflecting the changes in flaB mRNA detection, and further demonstrating that flaB mRNA levels can be used to follow spirochete burden in infected mice (12). bbk32 ands bbk50 mRNA levels were not significant on day 1, consistent with previous studies that B. burgdorferi N40 does not immediately express these genes following intradermal syringe challenge. Low levels of bbk50 were detected at the local site of inoculation on days 4–14 and reached levels of up to 6.0 ± 2 femtograms/tissue specimen on day 5. As the number of total spirochetes declines at the local site of inoculation on day 14, more bbk50 mRNA is expressed per spirochete at this later interval than at earlier time points. bbk32 mRNA levels greatly increased from days 3 to 7 and then declined. The change in bbk32 expression is consistent with the study of Probert and Johnson (13) demonstrating the BBK32 is a fibronectin-binding protein and may therefore required during the initial stage of infection.

Gene expression in the skin, joints, heart, spleen, and bladder was then assessed during disseminated murine infection (Table I⇑). Skin specimens were obtained from the murine mid-posterior thorax, away from the site of inoculation. flaB mRNA was detected in all of the tissues and at all intervals, confirming that spirochetes were present at these locations and that B. burgdorferi mRNA had been successfully isolated. flaB mRNA levels increased from days 14 to 30 and then decreased at 60 days, consistent with studies suggesting that the number of spirochetes decrease in mice as the host immune response to B. burgdorferi expands. At 14, 30, and 60 days, flaB mRNA levels were generally greater in the joints, heart, and spleen than in the skin and bladder, suggesting that the spirochete burden is higher in these tissues. bbk32 and bbk50 mRNA levels were lower than flaB mRNA levels at each location. At 14 days, bbk32 and bbk50 mRNA levels were generally similar in the tissues examined. At 30 days, bbk32 mRNA levels in the skin, heart, and spleen were greater than the bbk50 mRNA levels, particularly in the spleen. bbk32 and bbk50 mRNA levels in the joints and bladder at 30 days were not different. At 60 days, bbk32 and bbk50 mRNA levels declined as the number of spirochetes decreased in all tissues; however, spirochetes in the spleen continued to produce more bbk32 than bbk50 mRNA.

B. burgdorferi bbk32 and bbk50 expression in ticks

Previous research demonstrated that B. burgdorferi N40 expressed bbk32 and bbk50 during mammalian infection but not when cultured in vitro (11). B. burgdorferi are, however, maintained in a natural vector-host-vector cycle and not usually exposed to laboratory medium. bbk32 and bbk50 expression in ticks was therefore investigated to further understand when these genes are normally expressed. Individual spirochetes within ticks are readily discernible by IFA, unlike B. burgdorferi within murine tissues. Therefore IFA rather than RT-PCR was primarily used to examine spirochetes within the vector. BBK32 and BBK50 Ag on spirochetes within flat and engorged I. scapularis were examined by immunofluorescence using BBK32 or BBK50 antisera. BBK32 and BBK50 were not evident on spirochetes within unfed ticks but were apparent on B. burgdorferi in engorged ticks (Fig. 3⇓). The number of BBK32- and BBK50-producing spirochetes was determined by probing B. burgdorferi within flat and fed ticks with selected antisera. Fields containing ∼1000 spirochetes were first examined by dark-field microscopy. As the number of B. burgdorferi within ticks increases exponentially during engorgement, a larger number of fields from flat tick extracts were required to identify 1000 organisms. Following engorgement, 57 ± 3% of B. burgdorferi synthesized BBK32 and 78 ± 4% produced BBK50 (Fig. 4⇓). As expected, B. burgdorferi- or FlaB antisera (controls) readily identified spirochetes in flat and engorged ticks. OspA-producing spirochetes were only abundant in the flat ticks, and OspC-synthesizing organisms increased in number during feeding, consistent with previous reports (4, 5, 7). RT-PCR substantiated the IFA results (Fig. 5⇓). bbk32 and bbk50 mRNA were detected in engorged, but not flat ticks, whereas flaB mRNA (control) was evident in both groups of ticks. These studies demonstrated that bbk32 and bbk50 are expressed on B. burgdorferi within feeding I. scapularis ticks.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Immunofluorescence of spirochetes within engorged (A, C, E, and G) or flat (B, D, F, and H) ticks probed with specific antisera. A and B, B. burgdorferi-antisera. C and D, BBK32 antisera. E and F, BBK50 antisera. G and H, GT antisera (control).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Number of spirochetes within flat or engorged ticks that can be detected with selected antisera. Antisera used: B. burgdorferi, BBK32, BBK50, FlaB, OspA, OspC, and GT (control). The means ± SD of three experiments are shown.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

RT-PCR to detect B. burgdorferi bbk32, bbk50, or flaB mRNA within flat (A) or engorged (B) ticks. Lane 1, Molecular mass marker; lane 2, flaB; lane 3, bbk32; and lane 4, bbk50. +, with reverse transcriptase; −, without reverse transcriptase.

Immunization with BBK32 antisera is protective against tick-borne B. burgdorferi infection: BBK32 and BBK50 antisera do not, however, clear spirochetes from persistently infected animals

BBK32 and BBK50 antisera have previously been shown to protect mice from experimental challenge with an intradermal inocula of spirochetes and also to clear B. burgdorferi from mice infected with spirochetes for 24 h (11). We have now performed studies to examine whether the postinfection immunity phenomenon extends to later intervals after infection (Table II⇓). Results were similar when 10² rather than 10⁴B. burgdorferi were used as the challenge inocula (data not shown). Partial eradication was observed when BBK32 and BBK50 antisera were administered up to 2 days after infection. At 4 and 8 days after infection, when spirochetes had begun to disseminate from the skin to distant organs, B. burgdorferi clearance was no longer evident. In addition, passive immunization with BBK32 and BBK50 antisera at these intervals did not influence the severity of disease in the B. burgdorferi-infected mice. Arthritis was graded on a scale from 0 to 3, with 1, 2 and 3 representing mild, moderate, and severe inflammation, respectively. Tibiotarsal joints from BBK32- and BBK50-immunized mice had mild to moderate arthritis (1.7 mean ± 0.6 SD, similar to control mice (1.5 mean ± 0.5 SD).

Our results have now demonstrated that B. burgdorferi express bbk32 and bbk50 within ticks during engorgement before transmission to the host. We therefore also studied whether BBK32 and BBK50 Abs could protect mice against tick-borne infection (Table II⇑). Passive transfer of murine BBK32 antisera to naive mice partially prevented B. burgdorferi infection transmitted by tick bite. Murine BBK50 antisera did not provide immunity and the combination of murine BBK32 and BBK50 antisera did not afford greater protection than murine BBK32 antisera alone. Although the passive transfer of murine BBK32 antisera was partially protective against B. burgdorferi infection when 4 ticks engorged on mice, protection was not obtained when 20 ticks were placed on each animal.

Influence of BBK32 and BBK50 antisera on B. burgdorferi within ticks

Spirochetes synthesize BBK32 and BBK50 within ticks during engorgement and BBK32 antisera provides partial protection against tick-borne infection. Therefore, the effects of BBK32 and BBK50 antisera on spirochetes within engorging ticks were examined. The number of spirochetes within ticks that fed on mice administered BBK32 or BBK50 antisera were assessed by IFA (Table III⇓). Spirochetes were abundant in nymphal ticks that fed on mice given GT antisera (control). B. burgdorferi within ticks that engorged on mice administered BBK50 antisera also had substantial numbers of spirochetes. Ticks that fed on mice given BBK32 antisera had markedly fewer spirochetes (90% less) and the combination of BBK32 and BBK50 antisera did not result in a greater reduction in spirochete number.

The effect of BBK32 antisera on the survival of spirochetes following the molt was then examined. Nymphs that had engorged on immunized mice were allowed to molt to the adult stage over several months. Adult ticks were then examined for spirochetes using IFA. B. burgdorferi were readily detected within adult ticks that had been exposed to GT or BBK50 antisera during nymphal engorgement. All 12 GT-exposed ticks had spirochetes. Seven of 11 ticks exposed to BBK50 antisera had spirochetes. Only 2 of 11 ticks exposed to both BBK32 and BBK50 antisera and 3 of 12 ticks exposed to BBK32 antisera had B. burgdorferi. In addition, adult ticks that had been previously exposed to BBK32 or BBK32 and BBK50 antisera had markedly fewer spirochetes than controls (Table III⇑). The number of spirochetes in adult control ticks was less than that in engorged control nymphs because B. burgdorferi increase in number during tick feeding.

We then performed studies to determine whether this reduction in B. burgdorferi due to BBK32 antisera could prevent spirochete transmission from ticks to a mammalian host. Guinea pigs were used for these experiments because adult ticks do not engorge on mice and guinea pigs can be infected with B. burgdorferi (38, 43). Four or five adult ticks that had been exposed to GT, BBK32, or BBK50 antisera during nymphal engorgement were allowed to feed to repletion on guinea pigs. The guinea pigs were then assessed for exposure to B. burgdorferi infection by immunoblot (38). Guinea pigs fed upon by GT (three of four animals) or BBK50 (three of three animals) antisera-exposed ticks, but not ticks exposed to BBK32 (zero of four animals) or BBK32 and BBK50 (zero of three animals) antisera had evidence of B. burgdorferi infection. These data demonstrate that BBK32 antisera inhibited spirochete transmission by adult ticks.

BBK32 and BBK50 antisera inhibits larval acquisition

BBK32 Abs play an important role in B. burgdorferi survival within ticks. Therefore, the effects of BBK32 and BBK50 antisera on the larval acquisition of spirochetes were examined because these Ags are synthesized within engorging ticks and during mammalian infection. SCID mice were infected with B. burgdorferi and at 14 days the animals were administered BBK32 and BBK50 antisera. Larvae were allowed to engorge on the mice and spirochete transmission from the host to larvae was examined by IFA (Fig. 6⇓). Larvae that engorged on control animals had large numbers of spirochetes. In contrast, ticks that fed on BBK32- and BBK50-immunized mice had reduced numbers of B. burgdorferi.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Number of spirochetes within larvae that engorged on B. burgdorferi-infected SCID mice that had been passively immunized with BBK32 and BBK50 antisera. Groups of five mice were infected with B. burgdorferi for 2 wk and then administered antisera. Four larval ticks were then allowed to feed to repletion on the mice, collected, and then dissected and examined for spirochetes by immunofluorescence.