Membrane Interaction of Ibuprofen with Cholesterol-Containing Lipid Membranes

Abstract

Deciphering the membrane interaction of drug molecules is important for improving drug delivery, cellular uptake, and the understanding of side effects of a given drug molecule. For the anti-inflammatory drug ibuprofen, several studies reported contradictory results regarding the impact of ibuprofen on cholesterol-containing lipid membranes. Here, we investigated membrane localization and orientation as well as the influence of ibuprofen on membrane properties in POPC/cholesterol bilayers using solid-state NMR spectroscopy and other biophysical assays. The presence of ibuprofen disturbs the molecular order of phospholipids as shown by alterations of the ²H and ³¹P-NMR spectra of the lipids, but does not lead to an increased membrane permeability or changes of the phase state of the bilayer. ¹H MAS NOESY NMR results demonstrate that ibuprofen adopts a mean position in the upper chain/glycerol region of the POPC membrane, oriented with its polar carbonyl group towards the aqueous phase. This membrane position is only marginally altered in the presence of cholesterol. A previously reported result that ibuprofen is expelled from the membrane interface in cholesterol-containing DMPC bilayers could not be confirmed.

Keywords: NMR; cholesterol; fluorescence; ibuprofen; membrane interaction; membrane properties.

Figure 1

Chemical structure of ibuprofen (top) with the assignment of the resolved proton signals (red) and orientation of the dipole moment (black arrow) and isosurface of the electrostatic potential (bottom).

Figure 2

²H-NMR chain order parameters plots of POPC-d₃₁ in the absence and in the presence of 10 mol% ibuprofen and/or 20 mol% cholesterol.

Figure 3

Normalized rate constants (k_P) for dithionite (left) and ascorbate (right) reflecting the permeation of the respective anion across the membrane in POPC and POPC/Chol (80/20, mol/mol) LUVs. k_P values determined in the presence of ibuprofen (molar ratio lipid/drug = 5:1) were normalized to those determined in the absence of the drugs. For the dithionite assay, vesicles contained 1 mM lipid and 0.5 mol% NBD-PC. For the ascorbate assay, the lipid concentration was 2.5 mM and 2 mol% SL-PC. The data represent the mean ± SE of ≥ 6 (dithionite) and 8 (ascorbate) independent experiments.

Figure 4

¹H MAS NMR spectra of POPC membranes (top) and POPC/cholesterol (80/20, mol/mol) (bottom) in the presence of 10 mol% ibuprofen. Resolved signals of ibuprofen and cholesterol are marked; the assignment of the POPC peaks (nomenclature see chemical structure of POPC above the spectra) was taken from [23]. HDO denotes the signal of residual water. The NMR spectra were measured at a MAS frequency of 6,000 Hz and a temperature of 30 °C.

Figure 5

Induced chemical shifts for the ¹H signals of POPC due to ring current effects in the presence of 10 mol% ibuprofen in a POPC (red) and in a POPC/cholesterol (80/20, mol/mol) membrane (blue).

Figure 6

NOESY cross-relaxation rates between the three resolved signals of ibuprofen and the respective molecular segments of POPC along the membrane normal in POPC (red) and in POPC/cholesterol (80/20, mol/mol) (blue) membranes in the presence of 10 mol% ibuprofen. Cross-relaxation rates for the positions marked by * were not analyzed because of a significant signal overlap of ibuprofen and POPC signals. For the ibuprofen signal at 1.8 ppm, only cross peaks for the chain region of POPC were clearly observed.

Figure 7

Sketch of the membrane position and orientation of ibuprofen in (A) a POPC membrane and (B) POPC/cholesterol membrane deduced from the NOESY data in this study. Cholesterol molecules are represented by the blue ellipses.