Directed antigen delivery as a vaccine strategy for an intracellular bacterial pathogen

Abstract

We have developed a vaccine strategy for generating an attenuated strain of an intracellular bacterial pathogen that, after uptake by professional antigen-presenting cells, does not replicate intracellularly and is readily killed. However, after degradation of the vaccine strain within the phagolysosome, target antigens are released into the cytosol for endogenous processing and presentation for stimulation of CD8(+) effector T cells. Applying this strategy to the model intracellular pathogen Listeria monocytogenes, we show that an intracellular replication-deficient vaccine strain is cleared rapidly in normal and immunocompromised animals, yet antigen-specific CD8(+) effector T cells are stimulated after immunization. Furthermore, animals immunized with the intracellular replication-deficient vaccine strain are resistant to lethal challenge with a virulent WT strain of L. monocytogenes. These studies suggest a general strategy for developing safe and effective, attenuated intracellular replication-deficient vaccine strains for stimulation of protective immune responses against intracellular bacterial pathogens.

Fig. 1.

Generation of cytoLLO Lm. (A) Schematic of LLO expression constructs. LLO is a 529-aa polypeptide containing a secretion signal sequence at the N terminus. Transcription of the hly gene encoding LLO initiates from the PrfA-regulated phly promoter. The cytoLLO expression construct contains a 23-aa deletion of the secretion signal sequence and maintains the native transcriptional and translational control elements. The locations of the PrfA-binding site (PrfA box), hly promoter (phly), ribosome-binding site (RBS), initiating methionine (Met), and secretion signal are indicated. (B) Cytoplasmic fractions or secreted proteins present in culture supernatants of WT SLCC-5764 (WT), LLO-negative DH-L377 (ΔLLO), and the cytoLLO DH-L1233 (cytoLLO) Lm strains were analyzed by Western blot with a monoclonal anti-LLO antibody. Lanes 1 and 4 (M) show the mobility of a mass marker with size given in kDa.

Fig. 2.

Intracellular growth of Lm in BM-MAC. (A) BALB/c BM-MAC were infected with WT (squares) or pulsed with ΔLLO (circles) or cytoLLO (triangles) Lm and the number of intracellular bacteria determined. Results are the means ± SD of one of two experiments performed in triplicate with similar results. (B–D) At the indicated times postinfection, BM-MAC infected with WT (B) or pulsed with ΔLLO (C) or cytoLLO (D) Lm were fixed, stained, and examined by light microscopy. (B) Filled arrows indicate primary infected BM-MAC with replicating cytosolic bacteria. Open arrows indicate neighboring BM-MAC infected by cell-to-cell spread of WT Lm. (C and D) Arrowheads indicate phagocytosed ΔLLO (C) or cytoLLO (D) Lm. (E) BALB/c or SCID mice were injected with WT (≈600 CFU) or cytoLLO (≈2 × 10⁸ CFU) Lm. At the times indicated postinjection, the numbers of splenic CFU were determined. Data represent the difference between the numbers of CFU per spleen and the injection dose at the indicated time point. Determination of CFU for SCID mice injected with WT Lm was performed only at day 3 postinjection.

Fig. 3.

Presentation of listerial antigen to CD8⁺ effector T cells. (A) C57BL/6 BM-MAC were pulsed with 100 nM SIINFEKL peptide, infected with WT Lm secreting OVA (sOVA), or pulsed with ΔLLO or cytoLLO Lm expressing cytoplasmic-localized OVA (cOVA) and used as APCs in B3Z T cell-activation assays. (B) Cocultures of APCs and B3Z cells in (A) were examined by light microscopy after staining with X-gal. The multiplicity of infection used is indicated below each panel. (C) TAP1^−/− BM-MAC were infected with WT Lm secreting OVA (sOVA) or pulsed with ΔLLO or cytoLLO Lm expressing cytoplasmic-localized OVA (cOVA) and used as APCs in B3Z assays. (D) BALB/c BM-MAC infected with WT or pulsed with ΔLLO or cytoLLO Lm were cocultured with LLO_91–99-specific effector T cells in ELISPOT assays. Results presented in A, C, and D are the means ± SD of one of three independent experiments performed in triplicate with similar results. Results presented in B are representative of one of three independent experiments with similar results.

Fig. 4.

Immunization with cytoLLO Lm primes Lm-specific effector T cells. (A) BALB/c mice were immunized with one or two doses of WT, ΔLLO, or cytoLLO Lm, and the frequencies of LLO_91–99- (dark bars) or p60_217–225- (open bars) specific effector T cells were determined by ELISPOT. Results are presented as the numbers of IFN-γ-secreting cells per 100,000 spleen cells and represents the means ± SD of one of five independent experiments performed in triplicate with similar results (n = 4 mice per group). (B) The clearance of peptide-pulsed, fluorescently labeled target cells was determined 18 h after injection in immunized mice relative to naïve animals. The response of a representative mouse evaluated 10 days after booster injection is shown. The mean responses of immunized mice evaluated 6 (C) or 10 (D) days after booster injection are also presented. Data represent the means ± SD (n = 2–4 mice per group). *, P = 0.000037; **, P = 0.0469. (E) BALB/c mice were immunized with one or two doses of WT, ΔLLO, or cytoLLO Lm. Mice were challenged with a 10-LD₅₀ dose of Lm strain 10403 and the numbers of splenic CFU determined. Data are presented as log₁₀ protection compared with the naïve challenged group, and results are the means ± SD of three independent experiments (n = 4 mice per group).