Membranes & Electrostatics
The remarkable ability of lipids to self-organize into supramolecular structures such as monolayers, bilayers, and vesicles is essential for the viability of cell based life. Self-organization arises from inherent properties of the lipid molecules and intermolecular interactions between lipids and water. These intermolecular interactions are nominally referred to as hydrophobic, hydrophilic, dipolar, hydrogen-bond, and electrostatic. It is the combination of these interactions that produce the unique characteristics of the assembled bilayers. One of the most important properties of a cell is the electrostatic potential across the membrane and at the surface, relative to that of bulk water. While the membrane surface potential has been recognized as an essential feature of communication between cells and other structures the potential inside the bilayer relative to bulk water, herein referred to as the inner membrane potential, has been shown to affect the transport of ions, the insertion of membrane proteins, and other transmembrane processes. Substances such as ethanol alter the intermolecular hydrogen bonds among lipid head groups allowing more water into the membrane; melatonin and many anesthetics may alter both the surface and inner membrane potentials as well as disrupt intermolecular interactions; lipophilic ions, many membrane probes such as phloretin, and protonophores also alter the inner membrane potential.
We propose limits to the contributions to the total inner membrane potential by lipid head group components and by the orientation of water. These assessments are based on electrostatic calculations that were an integral part of our explanation of the photogating of ionic currents across the lipid bilayer. Our purpose is to describe an easy, fast method to calculate and examine the electrostatic consequences of various membrane structures and orientations, not to get embroiled in controversies regarding membrane structure. We do show how variations in these membrane dimensions and structures affect the dipole potential of the membrane by using a range of values quoted in the literature. The averages of these ranges are used as the parameters to calculate the total membrane potential.
The total potential inside the membrane is then obtained by summing the contributions of the various lipid and water components. The continuum, exponentially varying dielectric method of calculating the membrane potential is very rapid compared to molecular mechanics calculations, yet produces realistic results.
A calculation of a 100x100 array every 0.1nm along the z axis (60 arrays) takes <1 minute of computer time on a 120MHz personal computer. The algorithm is available on diskette.
Calculated potentials due to zwitterionic head groups (zi); surface water (sw); intercalated water ( iw); and the carbonyl dipole moment (cd) are added to give the total inner membrane potential.
This methodology for calculating the inner membrane potential, which uses a realistic dielectric profile, is clearly useful for examining the relative contributions of various lipid features and water to this potential. These calculations are readily adapted and applicable to calculations of the binding of drugs and xenobiotics to membranes and the resulting changes in membrane energetics. Calculations using the “average” structural parameters are more realistic than fixed structure models since the membrane is a fluid, dynamical system. The experimentally determined membrane potential of ~ +280 mV sets limits on the relative contributions of the carbonyl, water, and zwitterionic dipoles, and is more realistic than the ~ +600 mV calculated by molecular dynamics which are also limited to ~200 ns by limitations in computer time. Essentially similar membrane potential profiles are found using our dipole calculation for the carbonyl groups and zwitterionic head groups when we use the all-atom partial charges and coordinates given by Peitzsch et al. Our treatment does not take into account heterogeneous domains of lipids or water molecules within the polar head group of the bilayer. Note the dependence on membrane thickness such that thicker membranes result in substantially higher potentials for given charge densities within the bilayer. Head group rotation and the fluidity of the membrane would argue against static water and lipid domains. The figures show that the membrane potential and field are highly dependent on the parameters used and very small structural changes in the bilayer have large affects on the potential, and can be viewed as an adaptive mechanism to perturbations by exogenous substances and forces.
The dynamic properties of the membrane and the water allow it to minimize the local interaction energies which then result in a relatively constant potential inside the membrane. For example, the addition of lipophilic, charged species as probes of the membrane potential and dielectric may in fact cooperatively alter the balance between the structure of the lipid and the water dipole. At low charge densities, this cooperative adaptation of the structure of the membrane to an electrostatic perturbation is on a local scale, rather than being a global effect of the membrane. The exponential dielectric attenuates the local charge effect. The above calculations emphasize our previous conclusion and reinforce that discrete charges must be used, rather than a Gouy Chapman smeared charge, when considering potentials inside membranes.