Mechanisms of holin-mediated transport of proteins across the bacterial cytoplasmic membrane

• Applicant: Professor Dr. Thomas Brüser
• Subject Area: Metabolism, Biochemistry and Genetics of Microorganisms
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 560834849

Lytic phages must hydrolyze the host cell wall at the end of their reproductive cycle to induce lysis and thereby release their progeny. To do this, they produce endolysins, which hydrolyze the cell wall, and holins, which allow the endolysins access to the cell wall. For that purpose, most holins form holes in the cytoplasmic membrane. There are numerous unrelated families of these “canonical” holins, some of which are also used by bacteria to secrete proteins such as toxins without lysis. It is currently assumed that holins form two-dimensional aggregates (rafts) that dissolve large areas of the cytoplasmic membrane at the required time point. Working on T4 holin, a model system for canonical holins, we unexpectedly found that major accumulation of holins in the membrane is not required for holin function. Instead, we found that smaller assemblies are sufficient to enable the release of endolysin into the periplasm and that positions of the cytoplasmic domain of the holin and therein especially in the N-terminal amphipathic helix are essential for this. AlphaFold 3 predicts a reasonable ring structure for T4 holin, which is consistent with our data, as the important amphipathic helix forms a hydrophilic hole in it. The proposed project now aims to clarify whether ring-dependent hole formation really occurs in this way. With directed crosslinking we will detect contacts within the ring and in particular those of the amphipathic helix. A combination with the “hole-suppressing” exchanges will clarify whether the amphipathic helix folds inwards during ring formation and thereby forms the hole. We will also look at two of the unusual holins that bacteria use for the non-lytic secretion of toxins. These holins could form an aqueous pore similar to other holins, but AlphaFold 3 does not predict meaningful structures in these cases. One reason could be that AlphaFold 3 does not take the biophysics of membranes into account and that especially interactions of the amphipathic helices may depend on the presence of the membrane. We will therefore also investigate the mechanism of these holins biochemically and molecular biologically. As these holins function without causing cell lysis, we also want to know whether their pore formation is only transient or not. A purification of the analyzed holins will be attempted in order to characterize their associations also by cryo-EM. A defined pore could also explain a substrate selectivity of holins, whereby the generally very high pI of the endolysins could be a selection criterion, which we will investigate. Taken together, this project will be of great importance for elucidating the currently not understood holin mechanisms.

• DFG Programme: Research Grants