Evaluation of RevA, a fibronectin-binding protein of Borrelia burgdorferi, as a potential vaccine candidate for lyme disease

Abstract

Previous studies indicated that the Lyme disease spirochete Borrelia burgdorferi expresses the RevA outer surface protein during mammalian infection. As an adhesin that promotes bacterial interaction with fibronectin, RevA appears to be a good target for preventive therapies. RevA proteins are highly conserved across all Lyme borreliae, and antibodies against RevA protein are cross-reactive among RevA proteins from diverse strains. Mice infected with B. burgdorferi mounted a rapid IgM response to RevA, followed by a strong IgG response that generally remained elevated for more than 12 months, suggesting continued exposure of RevA protein to the immune system. RevA antibodies were bactericidal in vitro. To evaluate the RevA antigen as a potential vaccine, mice were vaccinated with recombinant RevA and challenged with B. burgdorferi by inoculation with a needle or by a tick bite. Cultured tissues from all treatment groups were positive for B. burgdorferi. Vaccinated animals also appeared to have similar levels of B. burgdorferi DNA compared to nonvaccinated controls. Despite its antigenicity, surface expression, and the production of bactericidal antibodies against it, RevA does not protect against Borrelia burgdorferi infection in a mouse model. However, passive immunization with anti-RevA antibodies did prevent infection, suggesting the possible utility of RevA-based immunotherapeutics or vaccine.

Fig 1

Antisera against B. burgdorferi B31 RevA recognizes Rev proteins across the Lyme borreliae. (A) Coomassie brilliant blue-stained 12.5% acrylamide gel of recombinant Rev (rRev) proteins. The positions of molecular weight markers are indicated to the left of the gel. B31 is the type strain of B. burgdorferi (28). N40 is an B. burgdorferi strain isolated from a tick (47). Strain 297, originally isolated from cerebrospinal fluid from a Lyme disease patient with meningitis (48), has two separate copies of RevA (19). PBi is a European isolate of B. garinii. (B) Western blot of the gel in panel A with affinity-purified antibody to RevA from strain B31. Mass spectrometric analysis (University of Kentucky Center for Structural Biology Protein Core Facility) indicated that the higher-molecular-weight bands present in most lanes are also RevA, suggesting that RevA may form homomultimers. Note that the affinity-purified antibodies to RevA from strain B31 do not cross-react with human plasma fibronectin.

Fig 2

Antibodies to RevA in infected mouse serum. BALB/c mice were infected via tick bite with B. burgdorferi. RevA-specific IgG and IgM from three individual mice were measured by ELISA. (A) IgM diluted 1:100. Data represent the mean absorbance and standard errors from 3 replicates per time point. (B and C) IgG diluted 1:100 (B) and 1:1,000 (C). Data in all three panels represent the mean absorbance ± standard errors (error bars) from 3 replicates per time point.

Fig 3

RevA antibodies are bactericidal. B. burgdorferi (5 × 10⁶/ml) in BSK-II medium was treated with a 1:25 dilution of anti-RevA antiserum (19) or preimmune sera for 24 h. Fifty microliters from each tube was transferred to 450 μl fresh BSK-II medium plus 6% rabbit serum. Motile bacteria were enumerated in 10 random fields after 5 days by dark-field microscopy. Data are normalized to the percentage for the no-antibody control and represent the means plus standard errors from 3 independent counts for each condition. ∗, P < 0.001 by Student's t test assuming unequal variances.

Fig 4

RevA antibody prior to B. burgdorferi challenge. After RevA immunization and prior to B. burgdorferi challenge, IgG (A) and IgM antibodies against RevA (B) in serum samples was measured by ELISA. Data represent the means ± standard errors (error bars) from 1 experiment with eight mice. RevA, RevA only; Adj, adjuvant alone; +, positive control (confirmed infected mouse); −, negative control (naive mouse).

Fig 5

RevA antibodies in mouse serum postinfection. C3H/HEN mice were infected via tick bite with B. burgdorferi. Three weeks after the final boost, mice were bled from the saphenous vein, serum samples were collected, and RevA-specific IgG (A) and IgM (B) were measured by ELISA. Data represent the means ± standard errors from 6 wells per mouse (n = 7) per condition. RevA, RevA only; Adj, adjuvant alone; +, positive control (confirmed infected mouse); −, negative control (naive mouse).

Fig 6

Serum from vaccinated mice is not bactericidal. B. burgdorferi (5 × 10⁶/ml) in BSK-II medium was treated with a 1:25 dilution of antiserum from vaccinated mice or preimmune sera for 24 h. Fifty microliters from each tube was transferred to 450 μl fresh BSK-II medium plus 6% rabbit serum. Bacteria were enumerated after 5 days by dark-field microscopy with a Petroff-Hausser chamber. Data are normalized to the percentage for the no-antibody control and represent the means and standard errors from 3 independent counts for each condition.