Borrelia burgdorferi OspC protein required exclusively in a crucial early stage of mammalian infection

Abstract

This study demonstrates a strict temporal requirement for a virulence determinant of the Lyme disease spirochete Borrelia burgdorferi during a unique point in its natural infection cycle, which alternates between ticks and small mammals. OspC is a major surface protein produced by B. burgdorferi when infected ticks feed but whose synthesis decreases after transmission to a mammalian host. We have previously shown that spirochetes lacking OspC are competent to replicate in and migrate to the salivary glands of the tick vector but do not infect mice. Here we assessed the timing of the requirement for OspC by using an ospC mutant complemented with an unstable copy of the ospC gene and show that B. burgdorferi's requirement for OspC is specific to the mammal and limited to a critical early stage of mammalian infection. By using this unique system, we found that most bacterial reisolates from mice persistently infected with the initially complemented ospC mutant strain no longer carried the wild-type copy of ospC. Such spirochetes were acquired by feeding ticks and migrated to the tick salivary glands during subsequent feeding. Despite normal behavior in ticks, these ospC mutant spirochetes did not infect naive mice. ospC mutant spirochetes from persistently infected mice also failed to infect naive mice by tissue transplantation. We conclude that OspC is indispensable for establishing infection by B. burgdorferi in mammals but is not required at any other point of the mouse-tick infection cycle.

FIG. 1.

Diagram of ospC wild-type, mutant, and complementing loci. (A) Structures of the wild type (B31-A3, top) and the ΔospC::flgB_p-kan mutant (ospCK1, bottom). The scale bar shows the coordinates on cp26 in kilobases. Oligonucleotide binding sites used in constructing pBSV2G-ospC and screening for the ospC genotype are represented by arrowheads (5 or 7 [5′] and 6 or 8 [3′]; see Table 1). Probes used in Southern blot assays (Fig. 4B) are shown as lines below genes. (B) Structure of pBSV2G-ospC (not drawn to scale). The flgB_p-aacC1 fusion confers gentamicin resistance on B. burgdorferi and E. coli (11). ori, colE1 origin of replication; IR, inverted repeat from cp9; ORF1 to ORF3, open reading frames allowing plasmid replication in B. burgdorferi (47).

FIG. 2.

OspC production by various B. burgdorferi strains grown in vitro. Left, silver-stained gel showing similar amounts of lysate loaded; right, Western blot probed with antibodies recognizing FlaB and OspC. wt, A3; mut, ospCK1; mut + ospC, ospCK1/pBSV2G-ospC. The values on the left are molecular sizes in kilodaltons.

FIG. 3.

Confocal images of spirochetes within tick salivary glands after 72 h of feeding. Panels: A, ospCK1 spirochete; B, ospCK1/pBSV2G-ospC spirochetes. Spirochetes were detected by immunofluorescence with rabbit anti-B. burgdorferi antibody and Alexa 488-labeled anti-rabbit antibody. Salivary gland cell nuclei were counterstained with DRAQ5. Scale bar, 10 μm.

FIG. 4.

Serological responses to infection with various strains of B. burgdorferi and assessment of maintenance of pBSV2G-ospC in B. burgdorferi mouse reisolates. (A) Western blot analysis of sera from mice injected with 5 × 10³ spirochetes of various strains. Samples loaded on gels: E. coli, lysate of E. coli carrying cloning vector; E. coli + P39, lysate of E. coli carrying cloning vector encoding B. burgdorferi proteins P39 and P28; Bb, lysate of B. burgdorferi. Representative results obtained with serum from one mouse per inoculated strain are shown. wt, A3; mut, ospCK1; mut + ospC, ospCK1/pBSV2G-ospC. The values on the left are molecular sizes in kilodaltons. (B) PCR screening of ospC loci of three reisolates from mice injected with ospCK1/pBSV2G-ospC (mouse reisolates) and controls (−, no DNA; wt, A3 DNA; mut, ospCK1 DNA; mut + ospC, ospCK1/pBSV2G-ospC DNA). Primers 7 and 8 (Fig. 1 and Table 1), which amplify unique fragments from wild-type and mutant loci, as indicated on right of the panel, were used. (C) Southern blot assay of two B. burgdorferi mouse reisolate DNA samples and controls probed with the ospC gene. wt, DNA derived from wild-type B. burgdorferi; mut + ospC, DNA derived from in vitro-grown ospCK1/pBSV2G-ospC mutant bacteria; mouse reisolates, DNA of two ospCK1* reisolates from mice injected with ospCK1/pBSV2G-ospC mutant bacteria. The ospC probe does not hybridize to cp26 of ospCK1* because the ospC gene is completely deleted in this strain. The values on the left are molecular sizes in kilobase pairs.

FIG. 5.

Stability of pBSV2G and pBSV2G-ospC in B. burgdorferi during growth in vitro and in mice. Plotted is retention of plasmids by B. burgdorferi after growth in culture (circles), in wild-type (wt) mice (squares), and in SCID mice (triangles). In vitro, cultures were plated after 50 doublings without selection for plasmid retention. In wild-type mice, two tissue types per mouse from three mice per strain were cultured without selection for plasmid retention 6 weeks after inoculation with A3/pBSV2G, A3/pBSV2G-ospC, or ospCK1/pBSV2G-ospC bacteria. For testing of pBSV2G-ospC stability in SCID mice, four mice were infected and three tissue types per mouse were cultured 6 weeks after inoculation. Each symbol represents the number of colonies that retained pBSV2G or pBSV2G-ospC out of 24 colonies screened per culture or tissue type. Differences in plasmid retention during growth in wild-type mice between pBSV2G in A3 bacteria and pBSV2G-ospC in A3 or ospCK1 mutant bacteria were determined to be significant with an exact, two-sided Wilcoxon test, with P ≤ 0.01. The stability of pBSV2G-ospC in SCID mice was also significantly greater than in wild-type mice (P < 0.01) but not significantly different from that of pBSV2G in wild-type mice (P = 0.39).