Anthrax lethal toxin-mediated killing of human and murine dendritic cells impairs the adaptive immune response

Abstract

Many pathogens have acquired strategies to combat the immune response. Bacillus anthracis interferes with host defenses by releasing anthrax lethal toxin (LT), which inactivates mitogen-activated protein kinase pathways, rendering dendritic cells (DCs) and T lymphocytes nonresponsive to immune stimulation. However, these cell types are considered resistant to killing by LT. Here we show that LT kills primary human DCs in vitro, and murine DCs in vitro and in vivo. Kinetics of LT-mediated killing of murine DCs, as well as cell death pathways induced, were dependent upon genetic background: LT triggered rapid necrosis in BALB/c-derived DCs, and slow apoptosis in C57BL/6-derived DCs. This is consistent with rapid and slow killing of LT-injected BALB/c and C57BL/6 mice, respectively. We present evidence that anthrax LT impairs adaptive immunity by specifically targeting DCs. This may represent an immune-evasion strategy of the bacterium, and contribute to anthrax disease progression. We also established that genetic background determines whether apoptosis or necrosis is induced by LT. Finally, killing of C57BL/6-derived DCs by LT mirrors that of human DCs, suggesting that C57BL/6 DCs represent a better model system for human anthrax than the prototypical BALB/c macrophages.

Figure 1. FACS Profile of Immature Human MoDCs

MoDCs were derived from human peripheral blood monocytes. Expression of HLA-DR, CD11c, CD14, CD16, CD80, CD86, and DEC-205 was assessed by flow cytometry. The data were collected from two subjects and are representative of similar experiments. Filled histograms represent isotype-matched controls.

Figure 2. Killing of Human DCs by LT

(A) MoDCs from a representative human subject were treated with LT (500 ng/ml PA and 250 ng/ml LF) or 10 μM camptothecin (Campt.) as a positive control, and annexin V/PI staining was measured by flow cytometry 48 h post-LT exposure. (B) Percentages of annexin V-positive untreated and LT-treated MoDCs from five different subjects were determined by flow cytometry 48 h post-LT exposure. (C) MTT assay of LT or camptothecin-treated human MoDCs. Mean + standard deviation from three independent experiments are shown. (D) LT-treated human MoDCs show signs of apoptotic cell death, as analyzed by electron microscopy 48 h post-LT exposure. Bars: 1 μm. (E) Human MoDCs were TUNEL-positive 48 h post-LT exposure. Untreated and LT-treated MoDCs were subjected to TUNEL and Hoechst staining.

Figure 3. FACS Profile of BMDCs

(A) BMDCs were derived from a BALB/c mouse and analyzed on day 10. Expression of CD11b, CD11c, CD14, and CD80 was assessed by flow cytometry. The data are representative of four similar experiments in BALB/c and C57BL/6 mice. (B) BALB/c and C57BL/6-derived BMDCs were stimulated by LPS for 18 h, and CD86 expression was measured by flow cytometry. Filled histograms represent isotype-matched controls.

Figure 4. BALB/c- and C57BL/6-Derived DCs Differ in Their Response to LT

(A) BALB/c and C57BL/6-derived BMDCs were treated with LT (500 ng/ml PA and 250 ng/ml LF), and cell survival was determined by MTT assay. (B) Caspase-3 activation of LT-treated BALB/c and C57BL/6 DCs as determined by a colorimetric caspase-3 cleavage assay. A representative experiment is shown. (C) The caspase inhibitors Z-VAD-FMK (10 μg/ml) and BOC-D-FMK (40 μg/ml) prevent caspase-3 activation in LT-treated BALB/c BMDCs. (D) The caspase inhibitors Z-VAD-FMK (10 μg/ml) and BOC-D-FMK (40 μg/ml) do not prevent LT killing of BALB/c BMDCs as determined by MTT assay. (E) C57BL6, but not BALB/c DCs, were TUNEL-positive post-LT exposure. BALB/c and C57BL/6-derived BMDCs were treated with LT (500 ng/ml PA and 250 ng/ml LF), and were stained using a TUNEL reaction and a Hoechst counterstain. (F) BALB/c and C57BL/6-derived BMDCs were analyzed by electron microscopy 4 and 48 h post-LT exposure, respectively. Bars, 1 μm.

Figure 5. MAPKK Cleavage Kinetics and LT Susceptibility of BALB/c- and C57BL/6-Derived BMDCs

(A) MAPKK-3 cleavage occurs at similar rates in murine and human DCs treated with LT (500 ng/ml PA and 250 ng/ml LF), as determined by Western blot analysis using anti-MKK3 and anti-actin (control) antibodies. (B and C) Relative LT susceptibility of BALB/c- (B) and C57BL/6-derived (C) BMDCs. BALB/c and C57BL/6 BMDCs were subjected to PA (500 ng/ml) and varying concentrations of LF. After 4 h (BALB/c DCs) and 72 h (C57BL/6 DCs), cell viability was determined by MTT assays. Representative experiments are shown.

Figure 6. Proteasome Inhibitors Block LT-Mediated Killing of BALB/c-Derived DCs

Proteasome inhibitors MG132 (10 μM) (A), or Velcade (0.1 μM) (B) were added either simultaneously with LT or 1 or 2 h post-LT exposure, and cell viability was determined by MTT assay. Mean + standard deviation of four independent experiments are shown.

Figure 7. In Vivo Depletion of DCs and Loss of T Cell Activating Function Following LT Exposure

(A) Murine DCs and macrophages are killed by LT in vivo. Ten-week-old BALB/c mice were injected intraperitoneally with LT, and the percentage of DCs and macrophages in the spleen were determined by flow cytometry. A representative experiment is shown. (B) Specific depletion of DCs and macrophages in LT-treated BALB/c mice injected intraperitoneally with either PBS or LT (200 μg PA and 200 μg LF). Levels of splenic DCs (CD11c⁺, MHC class II⁺), macrophages (CD11b⁺, MHC class II⁺), and circulating B cells (B220⁺) and T cells (CD3⁺) were determined by flow cytometry. Percent changes are shown by number of cells derived from LT-treated BALB/c mice (three mice per time point), relative to PBS-treated control mice (three mice per control experiment).