Anthrax Protective Antigen:  Prepore-to-Pore Conversion

• Published in: Biochemistry (Easton) 1999-08, Vol.38 (32), p.10432-10441
• Main Authors: Miller, Carl J, Elliott, Jennifer L, Collier, R. John
• Format: Article
• Language: English
• Subjects: anthrax; anthrax toxin; antigens; Antigens, Bacterial - chemistry; Antigens, Bacterial - genetics; Antigens, Bacterial - metabolism; Antigens, Bacterial - physiology; Bacillus anthracis; Bacillus anthracis - chemistry; Bacillus anthracis - genetics; Bacillus anthracis - metabolism; Bacillus anthracis - pathogenicity; bacterial toxins; Bacterial Toxins - chemistry; Bacterial Toxins - genetics; Bacterial Toxins - metabolism; Biological Transport - drug effects; Biological Transport - genetics; Cholic Acids - pharmacology; Chymotrypsin - metabolism; Detergents - pharmacology; Hydrogen-Ion Concentration; Hydrolysis; Ion Channels - chemistry; Ion Channels - genetics; Ion Channels - metabolism; Ion Channels - physiology; Models, Biological; Models, Molecular; PA63 protein; protective antigen; Protein Structure, Secondary - drug effects; Sequence Deletion; Spectrometry, Fluorescence
• ISSN: 0006-2960; 1520-4995
• DOI: 10.1021/bi990792d

PA63, the active 63 kDa form of anthrax protective antigen, forms a heptameric ring-shaped oligomer that is believed to represent a precursor of the membrane pore formed by this protein. When maintained at pH ≥8.0, this “prepore” dissociated to monomeric subunits upon treatment with SDS at room temperature, but treatment at pH ≤7 (or with β-octylglucoside at pH 8.0) caused it to convert to an SDS-resistant pore-like form. Transition to this form involved major changes in the conformation of loop 2 of domain 2 (D2L2), as evidenced by (i) occlusion of a chymotrypsin site within D2L2 and (ii) excimer formation by pyrene groups linked to N306C within this loop. The pore-like form retained the capacity to bind anthrax toxin A moieties and cell surface receptors, but was unable to form pores in membranes or mediate translocation. Mutant PA63 in which D2L2 had been deleted was inactive in pore formation and translocation but, like the prepore, was capable of forming heptamers that converted to an SDS-resistant form under acidic conditions. Our findings support a model of pore formation in which the D2L2 loops move to the membrane-proximal face of the heptamer and interact to form a 14-strand transmembrane β-barrel. Concomitantly, domain 2 undergoes a major conformational rearrangement, independent of D2L2, that renders the heptamer resistant to dissociation by SDS. These results provide a basis for further exploration of the role of PA63 in translocation of anthrax toxin's enzymic moieties across membranes.