Allelic variation of the Lyme disease spirochete adhesin DbpA influences spirochetal binding to decorin, dermatan sulfate, and mammalian cells

Abstract

After transmission by an infected tick, the Lyme disease spirochete, Borrelia burgdorferi sensu lato, colonizes the mammalian skin and may disseminate systemically. The three major species of Lyme disease spirochete--B. burgdorferi sensu stricto, B. garinii, and B. afzelii--are associated with different chronic disease manifestations. Colonization is likely promoted by the ability to bind to target tissues, and Lyme disease spirochetes utilize multiple adhesive molecules to interact with diverse mammalian components. The allelic variable surface lipoprotein decorin binding protein A (DbpA) promotes bacterial binding to the proteoglycan decorin and to the glycosaminoglycan (GAG) dermatan sulfate. To assess allelic variation of DbpA in GAG-, decorin-, and cell-binding activities, we expressed dbpA alleles derived from diverse Lyme disease spirochetes in B. burgdorferi strain B314, a noninfectious and nonadherent strain that lacks dbpA. Each DbpA allele conferred upon B. burgdorferi strain B314 the ability to bind to cultured kidney epithelial (but not glial or endothelial) cells, as well as to purified decorin and dermatan sulfate. Nevertheless, allelic variation of DbpA was associated with dramatic differences in substrate binding activity. In most cases, decorin and dermatan sulfate binding correlated well, but DbpA of B. afzelii strain VS461 promoted differential binding to decorin and dermatan sulfate, indicating that the two activities are separable. DbpA from a clone of B. burgdorferi strain N40 that can cause disseminated infection in mice displayed relatively low adhesive activity, indicating that robust DbpA-mediated adhesive activity is not required for spread in the mammalian host.

Fig. 1.

Recombinant DbpA proteins bind to biotinylated GAGs. Microtiter wells were mock coated (with PBS) or coated with His-tagged DbpA_N40,DbpA_B31, DbpA_PBr, or DbpA_VS461 and probed with biotinylated heparin, dermatan sulfate, or chondroitin-6-sulfate (see Materials and Methods). After a washing step, bound GAGs were detected by ELISA, as indicated by the absorbance at 450 nm. Each bar represents the mean of four independent determinations ± the standard errors. Asterisks indicate that GAG bound to DbpA-coated wells significantly (P < 0.05 [Student t test]) better than to mock-coated wells.

Fig. 2.

DbpA alleles from diverse Lyme disease spirochetes are expressed on the surface of a nonadherent B. burgdorferi strain. (A) Triton X-114 extractions of B. burgdorferi strain B314 expressing the indicated DbpA allele were subjected to SDS-15% PAGE and immunoblotted with a mixture of antisera raised against DbpA_B31, DbpA_N40-D10/E9, DbpA_PBr, and DbpA_VS461. (B) Spirochetes expressing DbpA protein were digested with proteinase K (PK+) or incubated in PBS (PK−). Lysates from 5 × 10⁷ treated bacteria were separated by SDS–15% PAGE, and DbpA and FlaB proteins were identified by Western blotting with an individual antiserum against DbpA_N40-D10/E9, DbpA_B31, DbpA_PBr, or DbpA_VS461. The proteinase K sensitivity experiments were performed several times and Fig. 2 is representative of such studies. Proteinase K-resistant species in lysates of B314/pDbpA_B356 and B314/pDbpA_N40-D10/E9, indicated by asterisks, were consistently observed and may represent intracellular nonlipidated DbpA protein. Flagellin, a periplasmic protein, was immunoblotted as a control for the retention of outer membrane integrity. B. burgdorferi strain B314 expressing DbpA_B31 was analyzed in a separate experiment.

Fig. 3.

DbpA-mediated binding to cultured mammalian cells is cell type specific and variable among alleles. Radiolabeled B. burgdorferi strain B314 spirochetes expressing the indicated DbpA allele were added to wells containing 293 epithelial cells or to empty wells (media). Each bar represents the mean (± the standard error) of four independent determinations. B. burgdorferi strain B314 expressing DbpA_B31 was analyzed in a separate experiment in which the binding by B314 strains expressing DbpA_PBr or no DbpA (vector control), analyzed in parallel as controls, bound with an efficiency similar to that in the depicted experiment.

Fig. 4.

Independently derived clones expressing the same allele of DbpA display similar in vitro binding phenotypes. Radiolabeled B. burgdorferi strain B314 spirochetes expressing either DbpA_VS461 (top panel), DbpA_N40 (bottom panel), or vector control (inset panel) were added to wells containing 293 epithelial cells or to wells containing either no cells (media) or decorin, dermatan sulfate, or PBS (mock coated). Each bar represents the mean (± the standard error) of four independent determinations.

Fig. 5.

Decorin- and GAG-binding levels vary among DbpA alleles. Radiolabeled B314 spirochetes expressing the indicated DbpA allele were added to wells coated with increasing concentrations of decorin or dermatan sulfate, and the percentage of bacteria stably bound was determined. Analysis of B314 expressing DbpA_B31 was performed in a separate experiment in which the binding by B314 strains expressing DbpA_PBr or no DbpA, analyzed in parallel to the controls, bound with an efficiency similar to that in the depicted experiment. Each point represents the mean of four independent determinations, with the standard error omitted for clarity. At a concentration of 1.25 mg of dermatan sulfate/ml, DbpA_PBr bound to this substrate at a significantly higher percentage (P < 0.05) than both DbpA₂₉₇ and DbpA_B31, and the latter two strains bound at a significantly higher percentage than DbpA_B356, DbpA_N40, and DbpA_VS461. At a concentration of 1.25 μg of decorin/ml, DbpA_PBr bound at a significantly higher percentage (P < 0.05) than DbpA₂₉₇,DbpA_B31, and DbpA_VS461, and the latter three bound at a significantly higher percentage (P < 0.05) than DbpA_B356 and DbpA_N40.

Fig. 6.

The C-terminal region of DbpA_VS461 is required for binding to GAGs and mammalian cells. (A) The C-terminal amino acid sequences of DbpA proteins analyzed in the present study. (B) Strain B314 expressing wild- type (WT) or mutant DbpA_VS461 were subjected to proteinase K (PK+) or mock (PK−) digestion and immunoblotted for DbpA or flagellin, as described for Fig. 1. Radiolabeled B314 expressing the vector control or the indicated DbpA_VS461 derivative was added to wells with 293 epithelial cells or no cells (C), or dermatan sulfate, decorin, chondroitin-6-sulfate, or BSA (D). The percentage of bacteria bound was determined by scintillation counting. Each bar represents the mean (± the standard errors) of four independent determinations.