(A) Schematic map of the plasmid pAdv142. The antigen expression cassette contains the hly promoter, followed by its signal sequence and 417 amino acids of the N terminus of nonhemolytic protein tLLO fused to human PSA. The PSA protein lacks its secretory signal sequence. (B) Expression and secretion of tLLO-PSA fusion protein by the strains dal dat 142 (lanes 2 and 4) and dal dat ΔactA 142 (lanes 1 and 3). The protein extracts were prepared by precipitating culture supernatants with 10% Trypticase soy agar. The Western blot represents the expression of tLLO-PSA in the total proteins secreted by the dal dat 142 and dal dat ΔactA 142 strains. The blot was stained with a 1:1,000 dilution of both anti-LLO and anti-PSA primary antibodies.

In vivo clearance of the dal dat ΔactA 142 strain in WT (left) and GKO^−/− (right) mice (A) and the dal dat 142 strain in WT mice (B). Mice were immunized intraperitoneally with 2 × 10⁶ CFU of the dal dat 142 strain and 10⁸ CFU of the dal dat ΔactA 142 strain. The bacterial load in the organs, livers, and spleens were determined after days 1, 2, 3, 7, and 10 of immunization. Two mice from each group were sacrificed on different days, and the number of bacterial CFU was determined by plating several dilutions of the homogenized spleens and livers on BHI plates containing 100 μg/ml of streptomycin. The plates were incubated at 37°C for 24 to 48 h for bacterial growth. Columns, mean number of CFU from each mouse; bars, standard deviation (SD).

(A) In vitro stability of the plasmid pAdv142 in the dal dat ΔactA 142 strain after 60 generations. The in vitro stability was determined by subculturing the dal dat ΔactA 142 strain in selective (BHI) and nonselective (BHI plus 100 μg/ml d-alanine) media for 8 days at a 1:10,000 dilution at 37°C and 200 rpm. The average CFU ± SD was determined for each day after plating on selective and nonselective plates in triplicate. Streptomycin (100 μg/ml) was added to the culture medium to restrict the growth of any potential contaminant. (B) In vivo stability was examined by immunizing mice with 5 × 10⁷ CFU of the dal dat ΔactA 142 strain intravenously in the tail vein. The CFU were determined in the homogenized spleens after 24, 48, and 72 h. Viable CFU were determined after plating on both selective and nonselective media. No colonies were recovered at the time points of 48 and 72 h. Columns, mean number of CFU from each mouse; bars, SD. (C) Intracellular growth of the dal dat 142 (▴) and dal dat ΔactA 142 (▪) strains in murine macrophage cells J774A.1. The CFU count was determined by taking duplicate samples at the indicated time points, followed by titration on BHI plates. The data shown are representative of the average CFU at each time point. The experiment was repeated three times, and data are representative of one study. Open circles, the L. monocytogenes 10403S strain.

(A) Tumor regression study using TPSA23 as a transplantable tumor model. Three groups of eight mice were implanted with 2 × 10⁶ tumor cells on day 0, and two groups were immunized with 10⁷ CFU of the Lm-LLO-PSA vaccine (positive control) (▴) and 10⁸ CFU of the dal dat ΔactA 142 strain (⋄) on days 6, 13, and 20. Naïve mice (○) did not receive any treatment. Tumors were monitored weekly, and mice were sacrificed if the average tumor diameter was 14 to 16 mm. Each symbol in the graph represents the tumor size of an individual mouse. The experiment was repeated twice, and similar results were obtained. (B) Percent survival of the naïve mice and immunized mice at different days of the experiment. Solid line, naïve mice; dashed line, mice immunized with the dal dat ΔactA 142 strain; broken line, mice immunized with the Lm-LLO-PSA vaccine.

PSA-specific immune responses were examined by tetramer staining and intracellular cytokine staining for IFN-γ. Mice were immunized twice with 10⁷ CFU of Lm-LLO-PSA and 10⁸ CFU of the dal dat ΔactA 142 strain at 1-week intervals. For immune assays, spleens were harvested on day 6 after the boost. Spleens from three mice per group were pooled for this experiment. (A) PSA-specific T cells in the spleens of naïve, Lm-LLO-PSA-, and dal dat ΔactA 142 strain-immunized mice using PSA-specific tetramer staining. Cells were stained with mouse anti-CD8 (fluorescein isothiocyanate), anti-CD3 (Percp-Cy5.5), anti-CD62L (APC), and PSA tetramer-PE and analyzed by FACSCalibur. (B) Intracellular cytokine staining to detect the percentage of IFN-γ-secreting CD8⁺ CD62L^low cells in the naïve (bar 1), Lm-LLO-PSA- (bar 2), and dal dat ΔactA 142 strain (bar 3)-immunized mice after stimulation with 1 μM of PSA-specific H-2D^b peptide for 5 h. (C) PSA-specific cytotoxic responses in mice immunized with the Lm-LLO-PSA vaccine (•) and the dal dat ΔactA 142 strain (□). Mice were immunized twice with 10⁷ CFU of Lm-LLO-PSA and 10⁸ CFU of the dal dat ΔactA 142 strain at a 1-week interval, and spleens were harvested on day 6 after the boost. The pooled splenocytes from each group were stimulated in vitro for 5 days after incubating with PSA-expressing stimulators. The target cells EL4 were pulsed with PSA H-2D^b peptide, HCIRNKSVIL, and specific lysis was determined after incubating targets with different ratios of effector cells for 3 h. The experiment was repeated three times, and representative data are shown. Filled triangle, naïve mice; E/T ratio, effector/target ratio.

Effect of vaccination of different Listeria constructs on the presence of PSA-specific TILs (A) and Tregs (B) in the tumors and spleens. Mice were implanted with TPSA23 tumors embedded in Matrigel and immunized with the irrelevant L. monocytogenes strain, the Lm-LLO-E7 vaccine (bars 3 and 4), and the dal dat ΔactA 142 strain (bars 5 and 6) on day 7 and day 14. On day 20, tumors and spleens were harvested from naïve (bars 1 and 2) and immunized mice. The tumors were pooled in each group to obtain an adequate number of cells for the staining of TILs and regulatory T cells. The cells were stained with surface antibodies, CD8, CD3, anti-CD62L, and PSA tetramer. To determine the Treg population in both tumors and spleens, the cells were stained for CD3, CD4, CD25, and Foxp3 antibodies. The experiment was repeated three times, and representative data are shown.