Functional analysis of the carboxy-terminal domain of Bacillus anthracis protective antigen

Abstract

Protective antigen (PA) is the common receptor-binding component of the two anthrax toxins. We investigated the involvement of the PA carboxy-terminal domain in the interaction of the protein with cells. A deletion resulting in removal of the entire carboxy-terminal domain of PA (PA608) or part of an exposed loop of 19 amino acids (703 to 722) present within this domain was introduced into the pag gene. PA608 did not induce the lethal-factor (LF)-mediated cytotoxic effect on macrophages because it did not bind to the receptor. In contrast, PA711- and PA705-harboring lethal toxins (9- and 16-amino-acid deletions in the loop, starting after positions 711 and 705, respectively) were 10 times less cytotoxic than wild-type PA. After cleavage by trypsin, the mutant PA proteins formed heptamers and bound LF. The capacity of PA711 and PA705 to interact with cells was 1/10 that of wild-type PA. In conclusion, truncation of the carboxy-terminal domain or deletions in the exposed loop resulted in PA that was less cytotoxic or nontoxic because the mutated proteins did not efficiently bind to the receptor.

FIG. 1

Schematic representation of the mutagenesis of domain 4 of PA. The three-dimensional structure of the carboxy-terminal domain of PA (amino acids 592 to 735), according to Petosa et al. (21), is shown. We truncated the PA molecule by introducing a stop codon after the codon encoding arginine 592 or aspartate 608. Two in-frame deletions resulting in the removal of 9 and 16 amino acids (residues 711 to 721 and 705 to 722, respectively) were also created in the exposed loop of 19 amino acids (residues 703 to 722). Point mutations are indicated by dashed arrows; deletions are indicated by plain arrows.

FIG. 2

Trypsin treatment of mutant PA. Wild-type and mutant PA proteins were purified from B. anthracis supernatants and studied in native form (−) or after (+) cleavage by trypsin. Samples (2 μg) were subjected to electrophoresis in a sodium dodecyl sulfate–12% polyacrylamide gel and stained with Coomassie blue. The values on the left are molecular masses in kilodaltons.

FIG. 3

Cytotoxicity assay for mutant PA. Wild-type or mutated PA protein (from 1 × 2 × 10⁴ to 2 × 10⁴ ng/ml) was serially diluted into a 96-well plate containing RAW264.7 cells in the presence of a constant concentration of LF (2 × 10⁴ ng/ml). The cells were incubated for 3 h, and their viability was assessed by using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (9). The experiment was carried out at least three times for each mutant PA protein.

FIG. 4

Interaction of mutant PA proteins with LF. Purified LF protein was subjected to electrophoresis in a 10% nondenaturing polyacrylamide gel and transferred to a nitrocellulose membrane under nondenaturing conditions. Wild-type or mutated PA proteins (2 μg), untreated (−) or treated (+) with trypsin, were added. The filter was incubated for 16 h at 4°C, and the LF-PA63 complex was detected with anti-PA serum.

FIG. 5

Assay of binding of wild-type or mutant PA proteins to CHO-K1 cells. Various concentrations of wild-type or mutant PA proteins (from 10 to 1 × 10⁴ ng/ml) were incubated with CHO-K1 cells for 3 h at 4°C. The binding of PA proteins to cells was detected with anti-PA serum and a secondary antibody coupled to β-galactosidase. MUG was added as the substrate for β-galactosidase, and fluorescence was quantified with a fluorimeter. The experiment was carried out at least three times for each mutant PA protein.