Minitrough 2

Monomolecular layers were prepared and monitored essentially as described previously 9, [20] [21] [22] . The equipment used was a Minitrough II (KSV Instruments Ltd, Finland).
Between 10 and 20 µl of a CHCl 3 :MeOH (2:1) solution of JO (1 mM) was spread on an unbuffered aqueous surface (24000 mm 2 ) at room temperature (24 o C ± 0.5 o C). About 5 min.
were allowed for the evaporation of the solvent. Lateral surface pressure (π) was measured by the Wilhelmy plate method. Shear may influence isotherm determinations. This phenomenon was discarded by placing the Wilhelmy plate in two perpendicular directions observing no effects on the registered isotherms. Before each experiment the trough was rinsed and wiped with 70 % ethanol and several times with bidistilled water. The absence of surface-active compounds in the pure solvents and in the subphase solution (bidistilled water) was checked before each run by reducing the available surface area to less than 10% of its original value after 5 minutes were allowed for the adsorption of possible impurities that might have been present in trace amounts.

Monomolecular layers were prepared and monitored with a Minitrough II (KSV Instruments Ltd., Finland) (see Supplementary data and ref [21] . for a detailed description). Briefly, CHCl 3 :MeOH (2:1) solutions of pure compounds were spread on an unbuffered aqueous surface at 25 ± 1 • C. The interface was compressed at a 10 mm 2 /s constant compression rate. Lateral surface pressure () was measured by the Wilhelmy plate method and the surface potential ( V) registered simultaneously by the voltage-measuring system (vibrating plate method; KSV Instruments, Helsinki, Finland). Reproducibility was within ±0.01 nm 2 , ±1mN/m and ±1 mV for molecular area (A), and V, respectively.
The surface parameters A min (minimum A occupied by an amphipathic molecule in a monolayer at the closest molecular packing) and c (collapse pressure, the maximal corresponding to A min ), were obtained from -A isotherms.