Functional complementation of staphylococcal alpha-hemolysin fragments. Overlaps, nicks, and gaps in the glycine-rich loop.

The final steps in assembly of the lytic pore formed by staphylococcal alpha-hemolysin (alpha HL) involve the formation of a nonlytic oligomeric pore precursor, followed by the formation of a transmembrane channel. In this study, truncation mutants of alpha HL encompassing the NH2-terminal or COOH-terminal half of the polypeptide chain and all, part, or none of the central glycine-rich loop were obtained by in vitro, coupled transcription and translation of mutant plasmid DNAs. These polypeptides were unable to oligomerize upon or cause lysis of rabbit erythrocytes (rRBCs). Twenty-one pairs of the same truncation mutants constituting discontinuous alpha HL chains with overlaps, nicks, and gaps in the central loop were obtained by cotranslation. When incubated with rRBCs, many of the pairs were able to form hetero-oligomers with wild-type alpha-hemolysin (s-alpha HL) and most of these formed homo-oligomers in the absence of s-alpha HL. However, only members of a subset of these pairs were able to lyse the cells. The lytic combinations contained overlaps, nicks, or gaps, but only two pairs, with nicks between amino acid residues 128 and 129 and between 131 and 132 had hemolytic activities approaching that of the wild-type polypeptide. Active combinations could also be obtained by separately translating NH2- and COOH-terminal truncation mutants and then combining them. These findings suggest that the integrity of the central loop is of little significance for oligomer formation but that it is more important for the final step in pore assembly or alternatively for determining the correct structure of the conductive channel. Our findings disagree with previous reports of NH2- and COOH-terminal fragments with hemolytic activity and of the prevention of hemolysis by proteolytic cleavage in the central loop. This discord is attributed to experimental and interpretative ambiguities in the earlier protein chemistry. For example, we show that loss of hemolytic activity after treatment with trypsin is not due to cleavage after Lys-131, as previously proposed, but to the removal of a small NH2-terminal peptide through cleavage after Lys-8.