Hydrogen-Bond Networks near Supported Lipid Bilayers from Vibrational Sum Frequency Generation Experiments and Atomistic Simulations

• Published in: The journal of physical chemistry. B 2018-05, Vol.122 (18), p.4870-4879
• Main Authors: Doǧangün, Merve, Ohno, Paul E, Liang, Dongyue, McGeachy, Alicia C, Bé, Ariana Gray, Dalchand, Naomi, Li, Tianzhe, Cui, Qiang, Geiger, Franz M
• Format: Article
• Language: English
• ISSN: 1520-6106; 1520-5207
• DOI: 10.1021/acs.jpcb.8b02138

We report vibrational sum frequency generation (SFG) spectra in which the C–H stretches of lipid alkyl tails in fully hydrogenated single- and dual-component supported lipid bilayers are detected along with the O–H stretching continuum above the bilayer. As the salt concentration is increased from ∼10 μM to 0.1 M, the SFG intensities in the O–H stretching region decrease by a factor of 2, consistent with significant absorptive–dispersive mixing between χ(2) and χ(3) contributions to the SFG signal generation process from charged interfaces. A method for estimating the surface potential from the second-order spectral lineshapes (in the OH stretching region) is presented and discussed in the context of choosing truly zero-potential reference states. Aided by atomistic simulations, we find that the strength and orientation distribution of the hydrogen bonds over the purely zwitterionic bilayers are largely invariant between submicromolar and hundreds of millimolar concentrations. However, specific interactions between water molecules and lipid headgroups are observed upon replacing phosphocholine (PC) lipids with negatively charged phosphoglycerol (PG) lipids, which coincides with SFG signal intensity reductions in the 3100–3200 cm–1 frequency region. The atomistic simulations show that this outcome is consistent with a small, albeit statistically significant, decrease in the number of water molecules adjacent to both the lipid phosphate and choline moieties per unit area, supporting the SFG observations. Ultimately, the ability to probe hydrogen-bond networks over lipid bilayers holds the promise of opening paths for understanding, controlling, and predicting specific and nonspecific interactions between membranes and ions, small molecules, peptides, polycations, proteins, and coated and uncoated nanomaterials.