The study shows that the peptide Humanin can stick to negatively‑charged fat bubbles (liposomes) and fold into a specific sheet shape, which seems important for its brain‑protecting action. A super‑active version (S14G) is much more stable in plain water, which may help it work better when it finally meets the liposome. Even a version that doesn’t work (S7A) still binds the liposome, so the membrane interaction itself is a key part of how Humanin works.

We have recently shown that a 24 amino acid Humanin (HN) adopts an anti-parallel β-sheet structure in the presence of a negatively charged 1,2-dioleoyl-sn-glycero-3-phosphoglycerol (DOPG) and suggested a possibility that it interacts with lipid membranes and thereby exerts neuroprotective effects through the target cell surface receptors or the intracellular signaling molecules following membrane interaction events. The structures of two HN analogs, having either a S7A mutation or a S14G mutation, were examined under the identical conditions, as the S7A analog is inactive and the S14G analog is 1000-fold more active than the wild type HN. These analogs showed a secondary structure indistinguishable from the structure of HN in the presence of DOPG liposome, while unrelated peptides were disordered with and without DOPG. It thus appeared that HN and the analogs, regardless of the biological activities, have an ability to interact with DOPG liposome and form an anti-parallel β-sheet structure. While the wild type HN and the S7A and S14G analogs were largely disordered in buffer, the S14G analog showed greater stability as a disordered structure in the buffer at a physiological temperature, suggesting that it maintains the disordered structure presumably required for the interaction with the DOPG liposome and thereby greater neuroprotective activity.