In vivo visualization of bacterial colonization, antigen expression, and specific T-cell induction following oral administration of live recombinant Salmonella enterica serovar Typhimurium

Abstract

Live attenuated Salmonella strains that express a foreign antigen are promising oral vaccine candidates. Numerous genetic modifications have been empirically tested, but their effects on immunogenicity are difficult to interpret since important in vivo properties of recombinant Salmonella strains such as antigen expression and localization are incompletely characterized and the crucial early inductive events of an immune response to the foreign antigen are not fully understood. Here, methods were developed to directly localize and quantitate the in situ expression of an ovalbumin model antigen in recombinant Salmonella enterica serovar Typhimurium using two-color flow cytometry and confocal microscopy. In parallel, the in vivo activation, blast formation, and division of ovalbumin-specific CD4(+) T cells were followed using a well-characterized transgenic T-cell receptor mouse model. This combined approach revealed a biphasic induction of ovalbumin-specific T cells in the Peyer's patches that followed the local ovalbumin expression of orally administered recombinant Salmonella cells in a dose- and time-dependent manner. Interestingly, intact Salmonella cells and cognate T cells seemed to remain in separate tissue compartments throughout induction, suggesting a transport of killed Salmonella cells from the colonized subepithelial dome area to the interfollicular inductive sites. The findings of this study will help to rationally optimize recombinant Salmonella strains as efficacious live antigen carriers for oral vaccination.

FIG. 1

In vitro characterization of Salmonella strains expressing GFP (lanes 1) or GFP_OVA (lanes 2). (A) SDS-PAGE. (B) Immunoblot with an antibody to GFP. (C) Immunoblot with an antibody to ovalbumin. (D) Absorbance spectra of Salmonella lysates. (E) Flow cytometry of SL3261(pGFP_OVA) cultures at the logarithmic (log) and stationary (stat) growth phases. The inset shows an SDS-PAGE gel of both cultures, with the arrow marking GFP_OVA. Similar data were obtained for three independent experiments.

FIG. 2

Colonization and in situ antigen expression of orally administered live SL3261(pGFP_OVA) cells. Averages and SEMs of CFU values for three mice per data point are shown. (A) Colonization of Peyer's patches (○, ●) and mesenteric lymph nodes (▵, ▴). The open symbols represent all recovered Salmonella cells, whereas the closed symbols represent Salmonella cells that retained the plasmid as determined by plating on selective medium. Similar results were obtained for three independent experiments. p.i., postimmunization. (B) Total in situ expression of GFP_OVA in the Peyer's patches (○) and mesenteric lymph nodes (▵) as determined by two-color flow cytometry (see text for description). (C) Subepithelial dome of a Peyer's patch 7 days postimmunization. Bar, 50 μm. The dotted line represents the apical border of the follicle-associated epithelium. L, intestinal lumen. (D) Enlarged micrograph of a single infected cell in the subepithelial dome 7 days postinfection. Bar, 5 μm. Similar images were obtained for 10 mice at various time points postimmunization from two independent experiments.

FIG. 2

Colonization and in situ antigen expression of orally administered live SL3261(pGFP_OVA) cells. Averages and SEMs of CFU values for three mice per data point are shown. (A) Colonization of Peyer's patches (○, ●) and mesenteric lymph nodes (▵, ▴). The open symbols represent all recovered Salmonella cells, whereas the closed symbols represent Salmonella cells that retained the plasmid as determined by plating on selective medium. Similar results were obtained for three independent experiments. p.i., postimmunization. (B) Total in situ expression of GFP_OVA in the Peyer's patches (○) and mesenteric lymph nodes (▵) as determined by two-color flow cytometry (see text for description). (C) Subepithelial dome of a Peyer's patch 7 days postimmunization. Bar, 50 μm. The dotted line represents the apical border of the follicle-associated epithelium. L, intestinal lumen. (D) Enlarged micrograph of a single infected cell in the subepithelial dome 7 days postinfection. Bar, 5 μm. Similar images were obtained for 10 mice at various time points postimmunization from two independent experiments.

FIG. 3

Detection of transgenic ovalbumin-specific T cells in chimeric mice after immunization with a recombinant Salmonella strain. (A) Flow cytometry of adoptively transferred ovalbumin-specific tgTCR CD4⁺ T cells in Peyer's patches of a recipient 1 day after transfer. (B) Total number of ovalbumin-specific T cells in Peyer's patches after various time intervals postimmunization (p.i.) with SL3261(pGFP_OVA). Averages and SEMs of values for three to seven mice per data point from two independent experiments are shown.

FIG. 4

Activation of ovalbumin-specific T cells after immunization with a recombinant Salmonella strain. (A) CD69 expression and forward scatter of ovalbumin-specific T cells in Peyer's patches (PP) and mesenteric lymph nodes (mLN) after oral immunization with SL3261(pGFP_OVA). Similar results were obtained for three independent experiments. d, days. (B) CD69 expression and forward scatter (fsc) of ovalbumin-specific T cells in Peyer's patches 2 and 7 days (d) after oral immunization with SL3261(pGFP). (C) Fluorescence of CFSE-labeled ovalbumin-specific T cells prior to or 12 days (d) after oral immunization (p.i.) with SL3261(pGFP_OVA).

FIG. 5

Blast formation and in vivo division of ovalbumin-specific T cells after oral immunization with a recombinant Salmonella strain. p.i., postimmunization. (A) Fraction of large cells (gated as in Fig. 4A) among ovalbumin-specific T cells in the Peyer's patches after oral immunization with SL3261(pGFP_OVA) (●) or SL3261(pGFP) (○). Averages and SEMs of values for three to seven mice per data point are shown. The statistical significance of differences between SL3261(pGFP) and SL3261(pGFP_OVA) was analyzed with the t test (∗∗, P < 0.001). Similar results were obtained for three independent experiments. (B) Total number of large ovalbumin-specific T cells in the Peyer's patches after oral immunization with SL3261(pGFP_OVA). (C) Fraction of CFSE¹⁰ cells (gated as in Fig. 4C) among ovalbumin-specific T cells in the Peyer's patches (○, ●) and mesenteric lymph nodes (⧫) after immunization with SL3261(pGFP_OVA) (●, ⧫) or SL3261(pGFP) (○). Averages and SEMs of values for three mice per data point are shown. The statistical significance of differences between SL3261(pGFP) and SL3261(pGFP_OVA) was analyzed with the t test (∗, P < 0.05). Similar results were obtained for two independent experiments. (D) Dose-response relationship between in situ expression of GFP_OVA and the fraction of large cells (gated as in Fig. 4A) among ovalbumin-specific T cells in the Peyer's patches 7 days after oral immunization with 1.2 × 10⁹ (+), 6 × 10⁹ (○), or 4 × 10¹⁰ (×) SL3261(pGFP_OVA) cells. The dotted line represents background staining after immunization with 4 × 10¹⁰ SL3261(pGFP) cells. The statistical significance of differences between SL3261(pGFP) and SL3261(pGFP_OVA) was analyzed with the t test (1.2 × 10⁹ CFU, P < 0.05; 6 × 10⁹ CFU, P < 0.01; 4 × 10¹⁰ CFU, P < 0.001).

FIG. 6

Colocalization of Salmonella cells (black) and ovalbumin-specific T cells (red) in the Peyer's patches 8 days after oral immunization with SL3261(pGFP_OVA). (A) Overview at low magnification. u, lumen; d, subepithelial dome; f, follicle; i, interfollicular region. Bar, 200 μm. (B) Dome area at higher magnification. Bar, 50 μm. (C) Interfollicular area at higher magnification. Bar, 50 μm.