Myeloid differentiation antigen 88 deficiency impairs pathogen clearance but does not alter inflammation in Borrelia burgdorferi-infected mice

Abstract

The spirochete Borrelia burgdorferi causes acute inflammation in mice that resolves with the development of pathogen-specific adaptive immunity. B. burgdorferi lipoproteins activate innate immune cells via Toll-like receptor 2 (TLR2), but TLR2-deficient mice are not resistant to B. burgdorferi-induced disease, suggesting the involvement of other TLRs or non-TLR mechanisms in the induction of acute inflammation. For this study, we used mice that were deficient in the intracellular adapter molecule myeloid differentiation antigen 88 (MyD88), which is required for all TLR-induced inflammatory responses, to determine whether the interruption of this pathway would alter B. burgdorferi-induced disease. Infected MyD88(-/-) mice developed carditis and arthritis, similar to the disease in wild-type (WT) mice analyzed at its peak (days 14 and 28) and during regression (day 45). MyD88(-/-) macrophages produced tumor necrosis factor alpha only when spirochetes were opsonized, suggesting a role for B. burgdorferi-specific antibody in disease expression. MyD88(-/-) mice produced stronger pathogen-specific Th2-dependent immunoglobulin G1 (IgG1) responses than did WT mice, and their IgM titers remained significantly elevated through 90 days of infection. Despite specific antibodies, the pathogen burden was 250-fold higher in MyD88(-/-) mice than in WT mice 45 days after infection; by 90 days of infection, the pathogen burden had diminished substantially in MyD88(-/-) mice, but it was still elevated compared to that in WT mice. The elevated pathogen burden may be explained in part by the finding that MyD88(-/-) peritoneal macrophages could ingest spirochetes but degraded them more slowly than WT macrophages. Our results show that MyD88-dependent signaling pathways are not required for B. burgdorferi-induced inflammation but are necessary for the efficient control of the pathogen burden by phagocytes.

FIG. 1.

Enhanced pathogen burden in B. burgdorferi-infected MyD88^−/− mice. B. burgdorferi flaB DNA was quantified in the urinary bladder of individual mice by real-time PCR. Each data point represents the result from an individual mouse normalized to the amount of host β-actin. At both 14 and 45 days after infection, urinary bladders from MyD88^−/− mice harbored significantly more DNA than those from WT mice (13-fold increase at day 14 [P = 0.0079 by the Mann-Whitney test] and 267-fold increase at day 45 [P = 0.0317 by the Mann-Whitney test]).

FIG. 2.

Reduced degradation of B. burgdorferi by MyD88^−/− macrophages. Adherent macrophages were exposed to spirochetes at a 1:10 ratio of macrophages to spirochetes, dipped in ddH₂O to eliminate extracellular spirochetes, and incubated in fresh medium until fixation at 0, 20, 60, 120, and 180 min as described in Materials and Methods. Monolayers were then fixed and double labeled for spirochetes (red) and the lysosomal marker LAMP-1 (green). Images were recorded on a Zeiss LSM 510 confocal imaging system using a 63× objective. (A) WT macrophages at 2 h. (B) MyD88^−/− macrophages at 2 h. (C) Percentages of elongated B. burgdorferi spirochetes associated with macrophages as a function of time. Note the protracted presence of elongated spirochetes in MyD88^−/− macrophages. The results are representative of two experiments.

FIG. 3.

MyD88^−/− mice exhibit a shift toward Th2-associated B. burgdorferi-specific antibody isotype responses. B. burgdorferi-specific antibody titers in the sera of WT and MyD88^−/− mice infected for 14 days (A) or 28 days (B) were measured by isotype-specific ELISA. *, the value for MyD88^−/− mouse sera differed significantly from the WT value for the indicated antibody class and isotype (P value range, 0.0007 to 0.05, by the Mann-Whitney test).

FIG. 4.

Western blot reactivities of sera from B. burgdorferi-infected MyD88^−/− and WT mice. Sera from individual mice infected for 28 days were assessed for reactivity to B. burgdorferi proteins by immunoblotting of B. burgdorferi lysates. (A) IgM reactivities of WT (lanes 1 to 5) and MyD88^−/− mouse sera (lanes 6 to 10). Arrows mark the locations of OspC and DbpA, which were assessed by the use of antisera for these proteins. (B) IgG reactivities of WT (lanes 3 to 7) and MyD88^−/− mouse sera (lanes 8 to 12). Lanes 1 and 2 depict the locations of OspA and OspC, respectively, as assessed by the use of OspA MAb VIIIC3.78 and an antiserum to recombinant OspC. The ∼30-kDa protein bound by antibodies in both WT and MyD88^−/− sera migrates below OspA. The unlabeled arrow marks the location of a protein targeted by the WT but not the MyD88^−/− immune serum. (C) Isotype-specific immunoblot of B. burgdorferi lysates using sera pooled from individual WT and MyD88^−/− mice. KO, MyD88^−/− mice.

FIG. 5.

Splenic CD4⁺-T-cell production of IL-4 and IFN-γ. CD4⁺ T cells were purified from the spleens of infected WT and MyD88^−/− mice and stimulated in vitro with B. burgdorferi lysates. After 72 h, supernatants were harvested and the amounts of IL-4 (A) and IFN-γ (B) were analyzed by cytokine-specific ELISAs. The results are reported as the means of triplicate samples ± standard errors of the means (A) or of duplicate samples (B). The minimum level of cytokine detection was 30 pg/ml for IL-4 and 30 pg/ml for IFN-γ.