Cationic amphipathic peptides (magainins, LL-37-class) adsorb flat on the outer leaflet at low peptide-to-lipid ratio (P/L). Above a critical P/L, enough peptides insert perpendicular to the bilayer that the lipids around them bend to line the hole with their own headgroups — a continuous curved monolayer connecting the two leaflets through the pore. This is the toroidal pore, distinct from a barrel-stave channel where only peptides (no lipid) form the wall.
The pore radius R obeys a line-tension vs. peptide-tension energy balance (classical pore nucleation theory, the same framework used for electroporation):
E(R) = 2πγR − πτR²
dE/dR = 0 → R* = γ/τ (critical / barrier radius)
τ = k·max(0, P/L − (P/L)crit(γ))
γ is the edge line tension (energy cost per unit length of exposed pore rim, ~5–30 pN for lipid bilayers); τ is the peptide-induced reduction of membrane tension, which grows with P/L once it clears a γ-dependent threshold. R* is an unstable equilibrium: a thermally-fluctuating pore with R < R* reseals (line tension wins); one that fluctuates past R* keeps growing (τ wins) — the same Kramers barrier-crossing picture used for electroporation and stochastic channel gating.
- P/L slider — raises peptide surface coverage, lowering the barrier and shrinking R*, exactly as in dose-response leakage assays.
- Line tension γ — models lipid composition (e.g. cholesterol raises γ, resisting pore opening and raising the P/L needed for lysis).
- Temperature — scales the random thermal kick added to R each step, so nucleation becomes a stochastic barrier-crossing event, not a deterministic switch.
- Nucleate seed pore — manually gives R a small kick, letting you see directly whether the current γ, P/L put R* above or below that seed (reseal vs. runaway growth).
Leakage flux is shown proportional to pore cross-section (∝ πR²), matching how AMPs collapse transmembrane ion gradients and leak cytoplasmic contents — the basis of their bactericidal action.