Anthrax toxin triggers endocytosis of its receptor via a lipid raft-mediated clathrin-dependent process

Abstract

The protective antigen (PA) of the anthrax toxin binds to a cell surface receptor and thereby allows lethal factor (LF) to be taken up and exert its toxic effect in the cytoplasm. Here, we report that clustering of the anthrax toxin receptor (ATR) with heptameric PA or with an antibody sandwich causes its association to specialized cholesterol and glycosphingolipid-rich microdomains of the plasma membrane (lipid rafts). We find that although endocytosis of ATR is slow, clustering it into rafts either via PA heptamerization or using an antibody sandwich is necessary and sufficient to trigger efficient internalization and allow delivery of LF to the cytoplasm. Importantly, altering raft integrity using drugs prevented LF delivery and cleavage of cytosolic MAPK kinases, suggesting that lipid rafts could be therapeutic targets for drugs against anthrax. Moreover, we show that internalization of PA is dynamin and Eps15 dependent, indicating that the clathrin-dependent pathway is the major route of anthrax toxin entry into the cell. The present work illustrates that although the physiological role of the ATR is unknown, its trafficking properties, i.e., slow endocytosis as a monomer and rapid clathrin-mediated uptake on clustering, make it an ideal anthrax toxin receptor.

Figure 1.

Proteolytic processing of PA triggers partitioning of the anthrax toxin into lipid rafts. (A) Wild-type CHO cells were incubated for 1 h at 4°C with 500 ng/ml of either a mixture of native and trypsin-nicked PA83 (PA63), or PA^SNKE. DRMs were prepared and analyzed by Western blotting against PA and caveolin-1. The load (L) corresponds to 1/10 of the total material on the gradient. (B) Wild-type, sphingomyelin-deficient (CHO SPB), and recomplemented (CHO SPB(lcb1)) CHO cells were treated or not with β-MCD or filipin, then incubated with 500 ng/ml PA83 for 1 h at 4°C followed by 30 min at 37°C. DRMs were prepared and fractions were probed for the presence of PA, caveolin-1, and flotillin-1 by Western blotting. (C) ATR-deficient CHO cells were stably transfected with human ATR having an HA-tag at the COOH terminus. DRMs were prepared and fractions were probed using a biotinylated anti-HA antibody. The two upper bands are endogenous biotinylated CHO cell proteins recognized by the streptavidin-HRP, even in cells not expressing ATR-HA. (D) CHO cells were incubated with 500 ng/ml of either trypsin-nicked PA83 or of PA^SNKE for 1 h at 4°C followed by 30 min at 37°C, washed at 4°C, and further incubated with 1 μg/ml LF for 1 h at 4°C before preparation of DRMs. Fractions were analyzed by Western blot with anti-LF pAb.

Figure 2.

Antibody cross-linking promotes raft association of PA83 in a cholesterol-dependent manner. (A) CHO cells were incubated at 4°C with 500 ng/ml PA83 for 20 min. Surface PA was then clustered by an antibody sandwich at 4°C. DRMs were prepared and fractions were probed for PA by Western blotting. (B) CHO cells were treated for 1 h at 4°C with 500 ng/ml PA83, and then were either fixed with PFA followed by incubation with primary and secondary antibodies to detect PA (Fix+Xlink), or incubated for 30 min at 4°C with rabbit polyclonal anti-PA, or for 30 min at 4°C with the secondary antibodies and then fixed (Xlink+Fix). Bar, 10 μm. (C) CHO cells were treated or not with β-MCD or filipin, then treated as in A. DRMs were prepared and fractions were probed for the presence of PA as in A. (D) ATR-HA–expressing CHO cells were incubated or not with 500 ng/ml PA83 for 20 min and submitted to antibody (polyclonal) clustering as in A. DRMs were prepared and fractions were probed for the presence of ATR-HA by HA antibodies. (E) CHO cells were treated for 1 h at 4°C with 500 ng/ml PA (either native PA83, trypsin nicked PA83, or PA^SNKE) and 500 ng/ml of an inactive aerolysin mutant (ASSP), a monovalent probe for GPI-anchored proteins. Cells were then successively incubated for 30 min at 4°C with rabbit polyclonal anti-PA and chicken polyclonal anti-aerolysin antibodies for 30 min at 4°C with corresponding fluorescent secondary antibodies. Cells were fixed and visualized. For ease of analysis, only a small region of the plasma membrane is shown for each condition. Bar, 2.5 μm. (F) CHO cells were treated as in B (Xlink+Fix), and were then permeabilized with saponin and double labeled for the presence of caveolin-1. Bar, 5 μm. (G) Schematic behavior of ATR at the cell surface. ATR and the ATR–PA83 complex are present in the glycerophospholipid area of the plasma membrane. On clustering of ATR either via heptamerization of PA63 or by anti-body cross-linking, the transmembrane protein is stabilized within lipid rafts.

Figure 3.

Raft association of PA triggers cellular uptake. (A) Antibody cross-linking leads to the appearance of intracellular PA83. As schematized in the right panel, CHO cells were incubated for 20 min at 4°C with 500 ng/ml PA^SNKE. Cells were then either (1) submitted to antibody cross-linking at 4°C and then further incubated at 4°C (condition: X-link+4°C) or 37°C (condition: X-link+37°C) for 30 min or (2) incubated at 37°C for 30 min and then submitted to the antibody sandwich at 4°C (condition: 37°C+X-link). For all conditions, cells were submitted or not to an acid wash before fixation. Bar, 10 μm. (B) Cholesterol depletion inhibits intracellular accumulation of PA63. CHO cells were treated or not with β-MCD, then incubated for 20 min at 4°C with 500 ng/ml PA^SNKE followed by an antibody sandwich (X-link+37°C). Internalization was allowed to proceed for 30 min at 37°C and cells were then submitted to a cold acid wash, fixed, and visualized using a fluorescent microscope.

Figure 4.

Disruption of lipid rafts blocks formation of the PA channel and intracellular delivery of LF. (A) CHO cells were treated or not with β-MCD, incubated for 1 h at 4°C with 500 ng/ml trypsin-nicked PA83, and were transferred to 37°C for different periods of time. Aliquots of 80 μg total cell extract proteins were loaded on a 7.5% SDS-gel and probed for PA by Western blotting. (B) CHO cells were treated or not with β-MCD, incubated for 1 h at 4°C with a mixture of 1 μg/ml trypsin-nicked PA83 and 1 μg/ml LF, and transferred to 37°C for different periods of time. 40 μg of total cell extracts were analyzed by Western blotting (12.5% SDS-gel) for the presence of LF, total MEK1 (anti-COOH–terminal antibody), LF-processed MEK1 (anti-NH₂–terminal antibody), and PA (using 7.5% SDS-gels for the latter).

Figure 5.

The anthrax toxin enters preferentially via a clathrin-mediated pathway. (A) HeLa cells were transiently transfected with the dominant-negative caveolin-1 mutant (GFP-Cav1), incubated for 1 h at 4°C with1 μg/ml trypsin-nicked PA83, and transferred to 37°C for different periods of time. 40 μg of total cell extracts were analyzed by Western blotting for the presence of heptameric PA63. (B) CHO cells were transiently transfected with caveolin-1-GFP (which has a wild-type phenotype), then incubated for 20 min at 4°C with 500 ng/ml PA^SNKE followed by an antibody sandwich as in Fig. 3 (X-link+37°C). Internalization was allowed to proceed for 30 min at 37°C, and cells were then fixed and visualized using a fluorescent microscope. A blow-up of a region of the plasma membrane is shown. Bar, 10 μm. (C) CHO cells were incubated for 20 min at 4°C with 500 ng/ml PA^SNKE followed by an antibody sandwich as in Fig. 3 (X-link+37°C). Internalization was allowed to proceed for 0, 5, 15, or 30 min at 37°C, cells were then submitted to a cold acid wash, fixed, and visualized. Bar, 10 μm. (D) CHO cells were incubated for 1 h at 4°C with1 μg/ml trypsin-nicked PA83, transferred to 37°C for different periods of time. 40 μg of total cell extracts were analyzed by SDS-PAGE and Western blotting for the presence of SDS-resistant heptameric PA63. The band intensity was quantified by densitometry (expressed in arbitrary units, a.u.), and is shown as a histogram to clearly illustrate the appearance and the degradation of the heptamer. (E) Cells were treated as in C, then incubated with protein A coupled to 10 nm gold, fixed, and processed for embedding in Epon and sectioning. Three examples are shown on which the clathrin coat is clearly visible (a–c, arrows); d is an example showing PA^SNKE in an invagination with no apparent coat (arrowhead). Bar, 200 nm. (F) CHO cells were incubated for 20 min at 4°C with 500 ng/ml PA^SNKE followed by an antibody sandwich (X-link+37°C). Internalization was allowed to proceed either for 5 min at 37°C in the presence of FITC-transferrin (Tf) or for 15 min in the presence of FITC-dextran. Cells were then submitted to a cold acid wash, fixed, and visualized. Examples of colocalization are indicated by arrowheads. Bar, 10 μm. (G) HeLa cells induced to express either wild-type or K44A dominant-negative dynamin-1 were treated and analyzed as in A. Numbers below lanes represent the incubation times at 37°C in min. The band intensities were quantified and plotted as in C. (H) HeLa cells transiently transfected with wild-type or dominant-negative EΔ95/295 Eps15 mutant were treated and analyzed as in G.