Analysis of the OspE determinants involved in binding of factor H and OspE-targeting antibodies elicited during Borrelia burgdorferi infection in mice

Abstract

Immune evasion by Lyme spirochetes is a multifactorial process involving numerous mechanisms. The OspE protein family undergoes antigenic variation during infection and binds factor H (fH) and possibly FHL-1/reconectin. In Borrelia burgdorferi B31MI, the OspE family consists of three paralogs: BBL39 (ErpA), BBP38, and BBN38 (ErpP). BBL39 and BBP38 are identical and therefore are referred to here as BBL39. The goals of this study were to assess the specificity of the antibody (Ab) response to the OspE paralogs and to identify the domains or determinants of OspE that are required for the binding of fH and OspE-targeting Abs that develop during infection. Here we demonstrate that at least some of the anti-OspE Abs produced during infection are paralog specific and that Ab binding requires conformational determinants whose formation requires both the N- and C-terminal domains of OspE. The binding of fH to OspE was also found to be dependent on conformational determinants. It is also demonstrated here that all of the OspE paralogs expressed by B. burgdorferi B31MI are capable of binding fH. The binding of fH to members of the OspF protein family was also assessed. In contrast to an earlier report, no binding of BBO39 or BBR42 to human fH was detected. Lastly, a series of competitive binding enzyme-linked immunosorbent assay analyses, designed to determine if fH and infection serum Abs bind to the same sites on OspE, revealed that these ligands interact with different regions of OspE.

FIG. 1.

Schematics of OspE (BBL39 and BBN38) constructs employed in this study and alignment of BBL39 and BBN38 amino acid sequences. (A and B) Schematics depict the BBL39 and BBN38 truncations that were generated and the nomenclature assigned to each. All constructs were generated as N-terminal S-Tag fusion proteins in E. coli as described in the text. Numbering is based on amino acid positions of the unprocessed proteins. (C) Alignment of the amino acid sequences of BBL39 and BBN38 (12).

FIG. 2.

Demonstration that a component of the anti-OspE Ab response elicited during infection in mice is paralog specific. E. coli harboring the full-length BBL39 or BBN38 construct was induced to express the r protein, and cell lysates were fractionated by SDS-PAGE and immunoblotted. Membranes were screened with sera collected from a mouse infected for 12 weeks with B. burgdorferi B31MIpc (A) or with infection sera preabsorbed with either BBL39 (B) or BBN38 (C) or both (D). All methods were as described in the text.

FIG. 3.

Analysis of the BBL39 and BBN38 determinants necessary for binding of anti-OspE Abs elicited during infection in mice. Subfragments of BBL39 and BBN38 were generated as S-Tag fusion proteins, fractionated by SDS-PAGE, immunoblotted, and screened with anti-rOspE antisera or 12-week infection sera. All methods were as described in the text. Note that addition of the S-Tag to each OspE subfragment adds approximately 17 kDa to the molecular mass.

FIG. 4.

Demonstration of the binding of infection-induced anti-OspE Abs to BBL39 subfragments fractionated by nondenaturing PAGE. Cell lysates of E. coli induced to express recombinant forms of BBL39 were fractionated by nondenaturing PAGE and immunoblotted as described in the text. Immunoblots were screened with anti-rOspE antisera or with infection sera as indicated.

FIG. 5.

Analysis of the binding of hfH to OspE and OspF proteins by use of an affinity ligand binding immunoblot format. r proteins were fractionated by SDS-PAGE and immunoblotted. As indicated, the blots were then incubated either with purified hfH or with blocking buffer prior to the addition of goat anti-fH antisera. All methods are described in the text. Asterisks indicate the migration position of FHL-1.

FIG. 6.

ELISA analysis of the abilities of recombinant BBL39 and BBN38 proteins or subfragments to bind polyclonal anti-OspE antisera and infection sera. Wells of ELISA plates were coated with OspE r proteins or subfragments. This and all other procedures are described in detail in the text. BBO39 served as a positive control for infection serum Ab binding, and BSA served as a negative control for both anti-rOspE antiserum and infection serum Ab binding. In one set of triplicate wells, anti-rOspE antisera were added (solid bars). Infection sera obtained from a mouse infected for 12 weeks with B. burgdorferi B31MIpc were added to another set of triplicate wells (open bars). Ab binding was measured as described in the text. Data are averages of three separate determinations. Error bars, standard deviations.

FIG. 7.

Analysis of the fH binding abilities of BBL39 and BBN38 subfragments. All methods were as described in the text. r proteins were immobilized in wells, and then one set of triplicate wells was incubated with a preparation of purified fH (solid bars) while a second set was incubated with blocking buffer (open bars). After a wash, either polyclonal anti-hfH antisera (A) or a monoclonal anti-hfH Ab (B) was added. Binding was measured as described in Materials and Methods and was expressed as the average of three determinations. Error bars, standard deviations.

FIG. 8.

ELISA analysis of binding of fH to BBL39 and BBN38 peptides. Wells of a 96-well microtiter plate were coated with peptides or appropriate controls as indicated. All procedures are described in the text. Data are averages of three separate determinations. Binding of peptides and r proteins to the wells was assessed by using anti-OspE antisera (solid bars). The abilities of the peptides to bind infection serum Abs were assessed by using anti-B. burgdorferi B31MIpc infection sera (hatched bars). Abilities to bind hfH were assessed by using anti-fH antisera (open bars).

FIG. 9.

Competitive binding ELISA analyses of fH and Ab binding to BBL39. All methods were as described in the text. In brief, BBL39 was immobilized in the wells of a 96-well microtiter plate in triplicate. Normal human serum (open bar) or purified hfH (shaded bar) was added, wells were washed, and infection sera (mouse) were added. Binding of mouse IgG to BBL39 was assessed by using rabbit anti-mouse IgG antisera. As a control for fH binding to immobilized BBL39, one set of wells was incubated with fH (solid bar) and washed, goat anti-fH antisera were added, wells were washed, and a rabbit anti-goat secondary Ab was added. To verify that anti-OspE IgG was able to bind to the immobilized BBL39, infection sera were added to one set of wells (crosshatched bar), wells were washed, and IgG binding was detected by using rabbit anti-mouse IgG. Detection methods were as described in the text. Data are average readings from three wells. Error bars, standard deviations.