(753c) Correlating Monolayer Molecular Structure With Interfacial Viscoelasticity

The surface viscosity of dipalmitoylphosphatidylcholine (DPPC) monolayers decreases by two orders of magnitude on addition of 3.7 mol% cholesterol, followed by a sharp increase in the monolayer elasticity above 3.7 mol%.  We correlate these intriguing rheological properties with changes in the molecular organization as revealed by Grazing Incidence X-ray Diffraction (GIXD).  Adding cholesterol at constant surface pressure decreases the tilt of the DPPC lattice in the same way as increasing the surface pressure at constant cholesterol. The correlation length of the DPPC lattice decreases with cholesterol fraction, suggesting that the increase in defect density leads to the decrease in surface viscosity.  Above 4 mol%, the DPPC lattice parameter and correlation length saturates, showing that the observed increase in monolayer elasticity comes about from a percolated, emulsion-like cholesterol network.

    Cholesterol is hypothesized to alter the fluidity, permeability and phase transitions of monolayers, bilayers, and cell membranes, but the mechanisms by which this happens are still obscure.  At different concentrations, cholesterol organizes phospholipid molecules into “liquid-ordered” phases; while at other concentrations, cholesterol inhibits phospholipid crystallization into solid or liquid condensed phases.  Despite their ubiquity in mammalian cell membranes, there is little quantitative information on the surface viscosity and elasticity of lipid/cholesterol monolayers, and how the viscosity and elasticity relate to interactions between cholesterol and phospholipids. For example, nanometer scale variations in membrane fluidity are hypothesized to be essential to the dynamics and function of membrane-associated proteins in self-organizing “rafts.” Monolayer fluidity also plays an important, but as yet unappreciated role in the spreading and surface tension lowering properties of lung surfactants.  Lung surfactant must reliably and reproducibly reduce the surface tension at the air-water interface of the alveoli to near zero to minimize the work of breathing. This requires fast spreading of the surfactant over the alveolar air-liquid interfaces during inspiration as the lung area rapidly increases.  During exhalation, surfactant must resist Marangoni flow due to the surface tension differences between the deep lung and the bronchi.   A lack of surfactant in premature infants causes potentially fatal neonatal Respiratory Distress Syndrome, which is currently treated with animal-derived replacement surfactants. However, the optimal concentration of many of the lipid and protein species, especially cholesterol is unknown.  Even the presence of cholesterol in replacement lung surfactants is controversial as blood and cell debris extracted with surfactant from animal lungs complicate the cholesterol analysis. FDA-approved clinical surfactants Survanta and Curosurf have all cholesterol removed, while FDA-approved Infasurf has 4-5 wt% cholesterol.  One common factor between all replacement surfactants is that DPPC is the dominant (50 – 80 mol%) phospholipid.

   Here we show that the rheological properties of mixed DPPC/Chol monolayers at low cholesterol  (≤ 3.7 mol%) arise from changes in the DPPC lattice, but the increased elasticity at higher cholesterol mole fractions is  the result of the mesoscopic evolution of the nanodomain morphology.   GIXD shows that increasing the Chol fraction at constant surface pressure, or increasing the surface pressure at constant Chol fraction decreases the tilt in the same way. DPPC tilts to accommodate the mismatch between the larger projected area of the phosphocholine headgroup relative to the alkane tailgroup at low surface pressure.  Palmitic acid (PA, used in Survanta) and hexadecanol (HD, used in the replacement surfactant, Exosurf) also decrease the molecular tilt at a given surface pressure. Cholesterol, like PA and HD, has a relatively larger tailgroup and a smaller headgroup, which mitigates the area mismatch, leading to the reduction in tilt. However, unlike PA or HD, the correlation length of the DPPC lattice decreases from ~ 10 nm to ~ 1 nm as Chol is increased from 0 to 7 mol%. This decreased correlation length is consistent with the cholesterol sterol ring disrupting the packing of the tailgroup lattice of DPPC.  The correlation length can be thought of as the distance between lattice defects; hence the reduction in lattice correlations with increasing cholesterol fraction suggest an increased defect density, which corresponds to an increase in the “free area” available to the molecules to enhance diffusion and hence, reduce the surface viscosity.  The reduction in lattice correlation and tilt saturates at ~ 4 mol% Chol, at the same concentration we see the change from a viscous to elastic monolayer (Fig. 4).  However, the reduction in correlations are not consistent with an increase in the elasticity of the monolayer, but rather suggests the opposite.  However, at mole fractions > 3.7%, the nanodomains percolate to form a bicontinuous morphology and the DPPC domain size decreases.  In analogy to 3-D surfactant-stabilized emulsions, even small stresses deform the emulsion network structure, leading to an elastic modulus, G’ ~ λR, in which λ, the line tension, is the two-dimension equivalent of the surface tension in 3-D emulsions. R is a characteristic length scale of the DPPC domains, which decreases with increasing Chol content as the nanodomains form a percolating network.