Autotransporter-Mediated Display of Complement Receptor Ligands by Gram-Negative Bacteria Increases Antibody Responses and Limits Disease Severity

Abstract

The targeting of immunogens/vaccines to specific immune cells is a promising approach for amplifying immune responses in the absence of exogenous adjuvants. However, the targeting approaches reported thus far require novel, labor-intensive reagents for each vaccine and have primarily been shown as proof-of-concept with isolated proteins and/or inactivated bacteria. We have engineered a plasmid-based, complement receptor-targeting platform that is readily applicable to live forms of multiple gram-negative bacteria, including, but not limited to, Escherichia coli, Klebsiella pneumoniae, and Francisella tularensis. Using F. tularensis as a model, we find that targeted bacteria show increased binding and uptake by macrophages, which coincides with increased p38 and p65 phosphorylation. Mice vaccinated with targeted bacteria produce higher titers of specific antibody that recognizes a greater diversity of bacterial antigens. Following challenge with homologous or heterologous isolates, these mice exhibited less weight loss and/or accelerated weight recovery as compared to counterparts vaccinated with non-targeted immunogens. Collectively, these findings provide proof-of-concept for plasmid-based, complement receptor-targeting of live gram-negative bacteria.

Keywords: autotransporter; complement; gram-negative; play; plug & tularemia; vaccine-targeting.

Figure 1

Conception and applicability of autotransporter (AT)—complement receptor ligand (CRL) fusions. (a) Graphical abstract. (b) The modification of YadA to yield YFC (YadA-FLAG-C3d) and YFP (YadA-FLAG-p28); SS: signal sequence. F: FLAG tag. (c) Heterologous expression of plasmid-borne YadA by Ec and Ft detected by western blot with α-YadA sera followed by horseradish peroxidase (HRP)-conjugated secondary Ab. Strains bearing empty vector are denoted by “-”. Arrows indicate the trimers and monomers of unmodified YadA. (d) Collagen-coated ELISA plates were incubated with serial dilutions of intact Ft:-, Ft:YadA, and Ft:YFC; bound bacteria were detected by α-Ft LPS Ab. (e,f) Whole cell lysates of Ec, Kp, and Ft containing empty vector (-) or the YFP or YFC expression vectors were probed by western blot with primary Ab specific for C3d and the FLAG epitope followed sequentially by biotinylated secondary Abs and streptavidin-HRP (SA-HRP). YFC and YFP trimers and monomers are designated with black and grey arrows. The ~20 kDa bands evident in all lanes are endogenously biotinylated bacterial proteins (annotated AccB in Ec and Ft [40]) detected by SA-HRP. (g) Intact bacteria as indicated were incubated in solution with α-FLAG Ab. (h) Intact Ft as indicated were incubated in solution with the indicated Ab. IglC is primarily a cytoplasmic Ft protein. Washed bacteria in (g,h) were probed for IgG heavy chain (HC), followed by biotinylated secondary Abs and SA-HRP.

Figure 2

Bacterial expression of YFC enhances binding and uptake by MΦs. SYTO-stained Ft strains (- and YFC) were incubated at various MOIs with RAW cells for 2 h at 4 °C to assess binding (a) or at 37 °C to allow for bacterial uptake (c). For microscopy images (insets), MOI = 100 and scale bars (bottom right) are 50 µm. Quantification of cell association was calculated as the SYTO signal bound to washed RAW cells divided by the total input SYTO signal (of unwashed Ft and cells) at each MOI. * t test p < 0.05. Results are combined from 4 independent experiments; means shown with standard deviation (SD). Un-stained Ft (- and YFC) were incubated with RAW cells at MOI = 0 (“no Ft”) or 500 for 2 h at 4 °C (b) or at 37 °C (d). Following washes to remove un-bound bacteria, cell-associated material was probed by western blot with Ab specific for the Ft protein FopA and the cellular protein β-actin.

Figure 3

YFC-bearing bacteria induce elevated p38 and p65 phosphorylation in MΦs. (a) RAW 264.7 cells were incubated without (“no Ft”) or with Ft:- or Ft:YFC (MOI = 500) at 37 °C for the indicated time. Washed cells were probed by western blot with Abs specific for the indicated proteins. (b,c) Densitometry of phosphorylated and total forms of p38 and p65 combined from ≥3 independent experiments. * t test p < 0.05.

Figure 4

Infection with YFC-bacteria induces higher titers of serum Ab. (a) Individual mouse weights recorded daily and expressed as a percentage of their baseline weight (average of d –3–0). 7–8 mice per group. # indicates t test p < 0.01 “-” vs YFC, low doses. * indicates t test p < 0.01 comparing the high doses. (b) Sera were collected on d21 post-vaccination and Ft-specific titers were determined via ELISA. Two-way ANOVA was used to assess the effect of YFC. 7–8 mice per group.

Figure 5

Vaccination with YFC-bacteria alters Ag recognition and limits challenge-induced morbidity. (a) Individual mouse weights recorded daily and expressed as a percent of their baseline weight (average of d –3–0); 8 PBS mice and 15–16 mice per vaccine group. * indicates t test p < 0.01 comparing Ft:- and Ft:YFC. (b) Sera collected 21 d post-vaccination (PV) were assessed by ELISA for Ft-specific Ig, IgG, and IgG2c titers. P values derived from t tests of “-” vs YFC. (c) PV sera pooled from “-” or YFC vaccinated mice were used at equal dilution to test IP with Ft Live Vaccine Strain (LVS). Washed bacteria were probed via western blot for Ig HC. (d) PV sera pooled from mice vaccinated with “-” or YFC were used at equal titers to probe by western blot aqueous (A), detergent (D), sarkosyl soluble (SS), and sarkosyl insoluble (SI) phases of WT Ft LVS. (e) PV sera from individual mice vaccinated with “-” or YFC were used at equal titers to probe by western blot A phases of WT Ft LVS. (f) The mice were challenged i.n. with 12k or 48k CFU of WT Ft LVS on d 28 PV. Individual post-challenge mouse weights recorded daily and expressed as a percent of their baseline weight (average of d 26–28 PV) with 4–6 mice per group. # t test p < 0.01 between “-” and YFC at the 12k dose. * t test p < 0.01 between “-” and YFC at the 48k dose. (g) Sera collected on d 21 PC were analyzed by ELISA along with PV sera for total Ft-specific Ig titers. * ANOVA with Tukey post-test p < 0.01 between indicated groups. Fold-increase (PC/PV) of average titer for each group is indicated in parentheses.

Figure 6

Mice vaccinated with YFC-bacteria display improved weight-gain following heterologous challenge. (a) Ft:- and Ft:YFC vaccinated mice were challenged i.n. with Ft LVS RML. Individual post-challenge mouse weights are expressed as a percent of their baseline weight (average of d 26–28 PV) with 7–8 mice per group. * t test p < 0.05 between “-” and YFC for 10⁵ challenge dose. (b) Sera collected on d 21 PC were analyzed by ELISA along with PV sera for total Ft-specific Ig titers. * ANOVA with Tukey post-test p < 0.05 between indicated groups. Fold-increase (PC/PV) of average titer is indicated in parentheses.