Tight cohesion between glycolipid membranes results from balanced water-headgroup interactions

Abstract

Membrane systems that naturally occur as densely packed membrane stacks contain high amounts of glycolipids whose saccharide headgroups display multiple small electric dipoles in the form of hydroxyl groups. Experimentally, the hydration repulsion between glycolipid membranes is of much shorter range than that between zwitterionic phospholipids whose headgroups are dominated by a single large dipole. Using solvent-explicit molecular dynamics simulations, here we reproduce the experimentally observed, different pressure-versus-distance curves of phospholipid and glycolipid membrane stacks and show that the water uptake into the latter is solely driven by the hydrogen bond balance involved in non-ideal water/sugar mixing. Water structuring effects and lipid configurational perturbations, responsible for the longer-range repulsion between phospholipid membranes, are inoperative for the glycolipids. Our results explain the tight cohesion between glycolipid membranes at their swelling limit, which we here determine by neutron diffraction, and their unique interaction characteristics, which are essential for the biogenesis of photosynthetic membranes.

Figure 1. Lipid structures and simulation set-up.

Chemical structures of a PC lipid (a) and of the glycolipid DGDG (b) as representatives of two fundamentally different lipid classes found in nature: Lipids with a headgroup chemistry dominated by one large electric dipole and lipids whose headgroups comprise multiple small electric dipoles in the form of OH groups. Both classes are schematically illustrated below the chemical structures. Dipoles are indicated by arrows. (c,d) Simulation snapshots of interacting DLPC and DGDG membranes, respectively, both at a large separation of D_w=2.3 nm. With periodic boundary conditions in all three directions, the simulations represent a periodic stack of membranes with adjustable hydration level. The simulation boxes are indicated with bright rectangles. For illustration, water molecules are only shown in the lower half of the box.

Figure 2. Comparison between experiments and simulations.

(a) Area per DGDG and PC lipid headgroup as functions of the water layer thickness D_w. Filled symbols indicate simulation results obtained with DGDG and with the PC lipid DLPC with semi-isotropic pressure coupling. Open symbols indicate experimental results for DGDG and DLPC membranes. Error bars for the simulation data represent 1 s.d. of uncertainty and were estimated from the scatter of the points around the plateau value at high hydration. For DLPC, they are smaller than the size of the symbols. The experimental error can be estimated in the same way as ⩽0.02 nm². (b) Pressure–distance curves of DGDG and PC lipid membranes as obtained in experiments (open symbols) and in the present simulations (filled symbols). Straight dashed lines in the semi-logarithmic plots indicate the best-matching exponential fits to the experimental data points for DGDG and to the combined experimental data sets for egg PC and DLPC. Error bars represent 1 s.d. of uncertainty.

Figure 3. Incorporation of the hydration water.

Density profiles of water, headgroups and hydrocarbon chains in hydrated DGDG (a) and PC lipid (b) membranes at D_w=0.23 nm. (c) Partial water volume v_w as a function of D_w for DGDG and PC lipids. Error bars represent 1 s.d. of uncertainty.

Figure 4. Hydrogen bonds.

(a) Total number of HBs per DGDG lipid as a function of the membrane separation. The dashed line indicates an exponential fit with decay length λ=0.12 nm. Inset: ll, lw and ww HBs. In the plot, ww refers to the excess HB number with respect to bulk water, −, see main text. Dashed lines indicate exponential fits with the same decay length λ=0.26±0.01 nm for all three curves. (b) The total number of HBs versus the interaction free energy for DGDG. Dashed line: Linear regression through all data points. (c) The same plot for PC lipids. Shaded region: High hydration limit where ≈0. The error bars in all three panels represent 1 s.d. of uncertainty.

Figure 5. Origin of the long-range repulsion.

(a) Density profiles of water and lipids in hydrated DGDG and PC lipid membranes at a large separation, D_w=2.3 nm. Vertical lines indicate the membrane surfaces at ±D_w/2. (b) Water orientation profiles 〈cos θ_w〉 at the same separation. Inset: Definition of the water dipole angle θ_w. (c) Distributions of DGDG and PC lipid headgroup orientations with respect to the membrane normal for large (D_w=2.3 nm, solid lines) and small (D_w=0.6 nm, dashed lines) separations. Inset: Definition of the headgroup vectors.

Figure 6. Dominant species of DGDG.

Chemical structures of dalDGDG (top) and palDGDG (bottom) lipids, differing in the saturation of the hydrocarbon chains.

Figure 7. Lamellar period at excess hydration.

Neutron diffraction of pure DGDG measured in excess D₂O between two silicon wafers. (a) 2D diffraction pattern. (b) 1D reduced curve. The insert is a plot of the Bragg reflection (h=1 to 3) positions versus h. The solid line is a linear fit to the experimental points yielding a period of 54.0±0.1 Å.