Identification of Surface Epitopes Associated with Protection against Highly Immune-Evasive VlsE-Expressing Lyme Disease Spirochetes

Abstract

The tick-borne pathogen Borrelia burgdorferi is responsible for approximately 300,000 Lyme disease (LD) cases per year in the United States. Recent increases in the number of LD cases, in addition to the spread of the tick vector and a lack of a vaccine, highlight an urgent need for designing and developing an efficacious LD vaccine. Identification of protective epitopes that could be used to develop a second-generation (subunit) vaccine is therefore imperative. Despite the antigenicity of several lipoproteins and integral outer membrane proteins (OMPs) on the B. burgdorferi surface, the spirochetes successfully evade antibodies primarily due to the VlsE-mediated antigenic variation. VlsE is thought to sterically block antibody access to protective epitopes of B. burgdorferi However, it is highly unlikely that VlsE shields the entire surface epitome. Thus, identification of subdominant epitope targets that induce protection when they are made dominant is necessary to generate an efficacious vaccine. Toward the identification, we repeatedly immunized immunocompetent mice with live-attenuated VlsE-deleted B. burgdorferi and then challenged the animals with the VlsE-expressing (host-adapted) wild type. Passive immunization and Western blotting data suggested that the protection of 50% of repeatedly immunized animals against the highly immune-evasive B. burgdorferi was antibody mediated. Comparison of serum antibody repertoires identified in protected and nonprotected animals permitted the identification of several putative epitopes significantly associated with the protection. Most linear putative epitopes were conserved between the main pathogenic Borrelia genospecies and found within known subdominant regions of OMPs. Currently, we are performing immunization studies to test whether the identified protection-associated epitopes are protective for mice.

Keywords: Borrelia burgdorferi; Lyme disease; VlsE; epitopes; immunization; protection; surface antigens; vls locus.

FIG 1

Repeated immunization assay. C3H mice of groups 1×, 3×, and 4× (6 animals per group) were immunized with the host-adapted ΔVlsE (ha-ΔVlsE) clone. At day 21 postimmunization, group 1× mice were challenged with ha-WT. Group 3× and 4× animals were also immunized with the ha-ΔVlsE clone at days 21 and 28. At day 35, group 3× mice were challenged with ha-WT and group 4× animals were additionally immunized with the ha-ΔVlsE clone. At day 42, the group 4× mice were challenged with ha-WT. Blood and other tissues (bladder, heart, tibiotarsal joint, and ear tissues) were sampled from each animal at days 7 and 21 after WT challenge, respectively. All tissues were cultured in BSK-II for a total of 4 weeks. The cultures were checked weekly for the presence of spirochetes by dark-field microscopy.

FIG 2

Western blots of 2-D gel electrophoresis gels of nonprotective (A) and protective (B) serum blotted against lysates of in vitro-grown wild-type B. burgdorferi B31. The whole-cell lysate of wild-type B. burgdorferi B31 (10⁶ cells per lane) was treated with nonprotective (A) and protective (B) mouse serum collected prior to challenge with the host-adapted wild type. The white ovals show the additional reactivity of the protective serum compared to that of the nonprotective serum. The blots can be compared to Coomassie blue-stained whole-cell lysates of B. burgdorferi B31 (see Fig. S1 in the supplemental material).

FIG 3

Identification of B. burgdorferi B31 proteins associated with a protective immune response in ΔVlsE clone-immunized C3H mice. Mouse serum was initially collected from the repeatedly immunized animals prior to ha-WT challenge. Upon completion of the repeated immunization assay, the samples were categorized as protective (4 samples) and nonprotective (3 samples) on the basis of the results of the challenge experiment. The sera were then analyzed via a random peptide phage display library in order to identify peptide (antibody) repertoires of protective (repertoire A) and nonprotective (repertoire B) sera. Repertoires A and B were then compared via the permutation test. As a result, 761 peptides (repertoire C) were found to be significantly associated with the protective sera. Out of the 761 peptides, 400 peptides were detected in all protective serum samples and absent in the 3 nonprotective serum samples. These 400 peptides were then mapped to the amino acid sequences of 1,391 proteins from the complete genome of B. burgdorferi B31 via BLASTP analysis. Only matches that had at least 4 exact amino acid matches were first considered, for a total of 1,516 hits. Then, 318 gapless hits with at least 5 contiguous identical amino acids only were considered, resulting in the identification of 254 proteins. A total of 94 peptides with hits to a single protein represented 85 out of 254 proteins.

FIG 4

Conformational epitopes (labeled A1 through A4 and B1) of B. burgdorferi B31 outer surface proteins A (OspA) and B (OspB) with aligned protection-associated peptides. Each epitope label is followed by its cognate monoclonal antibody (MAb) name. A1 (36), A2 (40), A3 (37), and A4 (39) are on OspA, while B1 (38) is on OspB. Epitope residues are underlined in the protein sequences, with representative sequence positions being numbered. Peptide residues are rendered in uppercase if they are part of a BLASTP hit alignment or lowercase otherwise. Peptide identifiers are in parentheses. Residues of A3 marked with asterisks are critical for LA-2 binding and complement-dependent bactericidal activity (95). Residues common to both A3 and A4 are marked with carets. A3 is topologically analogous to B1, with B1 being centered around OspB residue K253 (marked with a plus sign), which is essential for binding of OspB by H6831 (38). The other MAbs, MAb B3G11 and MAb N5G5, bind 10-mer peptide analogs (sequences highlighted with a magenta background) of OspA and OspB, respectively (41).

FIG 5

Protective sequences of B. burgdorferi B31 outer surface proteins A (OspA), B (OspB), and C (OspC) with aligned protection-associated peptides. Underlined sequences (with the N- and C-terminal residue positions being numbered) of OspA, OspB, and OspC elicit antipeptide antibodies with complement-dependent (anti-OspA and anti-OspC) and -independent (anti-OspB) bactericidal activity (42–44). Peptide identifiers are on the lines of their respective sequences, whose residues are rendered in uppercase if they are part of a BLASTP hit alignment or lowercase otherwise. BLASTP analysis E values and bit scores are in parentheses. OspA and OspB sequences are aligned to emphasize their homology, with OspB residue K253 being marked with a plus sign, as in Fig. 4 (within conformational epitope B1). A subsequence exactly matching OspC residues 131 to 149 (highlighted with a yellow background) forms part of a chimeric vaccinogen that elicits antipeptide antibodies having complement-dependent bactericidal activity (96).

FIG 6

B. burgdorferi B31 outer surface protein C (OspC) sequence with aligned epitope sequences (labeled E1 through E6) and protection-associated peptides. Epitope sequences (with the N- and C-terminal residue positions being numbered) were inferred from 12-mer phage-displayed peptide sequences of B. burgdorferi SKT-2 OspC (45). Homologous sequences in B. burgdorferi B31 OspC are underlined. Epitope sequence differences between the two B. burgdorferi strains are highlighted with a cyan background. Peptide residues are rendered in uppercase if they are part of a BLASTP hit alignment or lowercase otherwise. Peptide identifiers are in parentheses. Angled brackets delineate residues 130 to 150, which constitute the protective sequence that contains a vaccinogen component subsequence (highlighted with a yellow background). The 15-mer peptide whose sequence exactly matches residues 133 to 147 (highlighted with a magenta overhead band) is bound by bactericidal complement-independent MAb 16.22 (47).

FIG 7

Development of a protective antibody response against host-adapted B. burgdorferi upon repeated immunizations with the ha-ΔVlsE clone. (A to C) The diagrams show how a protective antibody response may have developed in mice repeatedly immunized with the ha-ΔVlsE clone. It is plausible that VlsE sterically shields mostly dominant epitopes of the B. burgdorferi surface from host antibodies. After a single immunization, antibodies against dominant epitopes (purple circles) of various surface antigens (green ovals) (A) are primarily developed. However, repeated immunizations with the ha-ΔVlsE clone result in the development of antibodies against previously subdominant epitopes (orange circles) (B). Upon challenge with VlsE-expressing B. burgdorferi B31 (ha-WT), the subdominant epitopes are fully accessible to antibodies, as VlsE does not supposedly shield them. Some of these subdominant epitopes may induce protective antibodies.