Holins are phage-encoded small transmembrane proteins that perforate the bacterial cytoplasmic membrane. In most cases, this process allows the phage-encoded peptidoglycan hydrolases to act on the cell wall, resulting in host cell lysis and phage release. We report a detailed functional characterization of Mycobacterium phage D29 gp11 coding for a putative holin that, upon expression, rapidly kills both Escherichia coli and Mycobacterium smegmatis. We dissected Gp11 by making several deletions and expressing them in E. coli. The shortening of Gp11 from its C-terminus results in diminished cytotoxicity and smaller holes. Evidently, the two transmembrane domains (TMDs) present at the N-terminus of Gp11 are incapable of integrating into the cytoplasmic membrane and do not show toxicity. Interestingly, the fusion of two TMDs and a small C-terminal region that bears the coiled-coil motif resulted in restoration of the cell killing ability of the protein. We further show that the second TMD is dispensable in protein toxicity because its deletion does not abolish Gp11-mediated cell death. We conclude that Gp11 C-terminal region is necessary but not sufficient for toxicity. These results shed light on a yet undiscovered role of Gp11 C-terminal region that will help clarify the mechanism of holin-mediated membrane perforation. Finally, we abolish the toxicity of Gp11 using a specific Gly to Asp substitution in the putative loop region of the protein; the mutant protein may help to clarify how holin functions in mycobacteriophage D29.

Bacterial cell lysis during bacteriophage infection is timed by perfect orchestration between components of the holin-endolysin cassette. In bacteria, progressively accumulating holin in the inner membrane, retained in its inactive form by antiholin, is triggered into active hole formation, resulting in the canonical host cell lysis. However, the molecular mechanism of regulation and physical basis of pore formation in the mycobacterial cell membrane by D29 mycobacteriophage holin, particularly in the nonexistence of a known antiholin, is poorly understood. In this study, we report, for the first time, the use of fluorescence resonance transfer measurements to demonstrate that the first transmembrane domain (TM1) of D29 holin undergoes a helix ↔ β-hairpin conformational interconversion. We validate that this structural malleability is mediated by a centrally positioned proline and is responsible for controlled TM1 self-association in membrana, in the presence of a proton gradient across the lipid membrane. We demonstrate that TM1 is sufficient for bacterial growth inhibition. The biological effect of D29 holin structural alteration is presented as a holin self-regulatory mechanism, and its implications are discussed in the context of holin function.