Membrane curvature induces cardiolipin sorting

Abstract

Cardiolipin is a cone-shaped lipid predominantly localized in curved membrane sites of bacteria and in the mitochondrial cristae. This specific localization has been argued to be geometry-driven, since the CL's conical shape relaxes curvature frustration. Although previous evidence suggests a coupling between CL concentration and membrane shape in vivo, no precise experimental data are available for curvature-based CL sorting in vitro. Here, we test this hypothesis in experiments that isolate the effects of membrane curvature in lipid-bilayer nanotubes. CL sorting is observed with increasing tube curvature, reaching a maximum at optimal CL concentrations, a fact compatible with self-associative clustering. Observations are compatible with a model of membrane elasticity including van der Waals entropy, from which a negative intrinsic curvature of -1.1 nm^-1 is predicted for CL. The results contribute to understanding the physicochemical interplay between membrane curvature and composition, providing key insights into mitochondrial and bacterial membrane organization and dynamics.

Keywords: Membrane biophysics; Research data.

Fig. 1

CL is enriched in curved membranes. a Schematic of tube assay. A bead in an optical trap is used to pull a membrane tube from a GUV held by a micropipette. The pressure in the micropipette controls the membrane tension and the resulting tube radius. The concentration of the reference lipid and the CL in the tube are measured by confocal fluorescence microscopy. b Confocal image of a tube pulled from a GUV containing CL. The membrane (red) was labelled with a fluorescent reference lipid (Bodipy TR-Ceramide) and the CL (green) with Top-Fluor CL. Contrast has been enhanced, and green and red intensities have been scaled to match in the GUV (which is orange-yellowish). The green color of the tube reflects CL enrichment in the tube (relative to the GUV). c Images and intensity profiles of tubes pulled from GUVs containing CL for large (R ≈ 37 nm) and small (R ≈ 10 nm) tube radii. CL is enriched in curved membranes obtaining higher intensity in the green channel compared to the red channel in small tubes. d Box plots comparing the sorting ratio for curved tubes (c = 0.10 ± 0.03 nm⁻¹) pulled from GUVs containing green fluorescent lipids: control (in orange) and a CL density of 0.10 ± 0.05 molecules per nm² (in green). The median is represented with a line; the box plot represents the 25th–75th percentiles; and the error bars show the 5th–95th percentile. CL is enriched in the tubes (average sorting ratio 1.9 ± 0.3, N = 10 GUVs) comparing with the lipid control (average sorting ratio 0.9 ± 0.2, N = 21 GUVs). (E) CL molecules bend the membrane in the direction of the imposed curvature (inner monolayer, which drives CL enrichment) whereas bends the membrane against the imposed curvature (outer monolayer, which causes CL depletion)

Fig. 2

CL enrichment as a function of CL density in GUVs. CL sorting as a function of the area fractioning the GUV ρ_GUV,CL × A_CL (in percentage) for four ranges of tube curvature: very high c = 0.13 ± 0.02 nm⁻¹; high 0.09 ± 0.02 nm⁻¹; low 0.062 ± 0.012 nm⁻¹; and very low 0.034 ± 0.00 nm⁻¹. The points are the arithmetic averages of binned CL GUV densities and sorting ratios, and the error bars represent the measurement error accumulated to the corresponding standard deviations. Dashed lines represent the minimum square fit to the non-interacting uncoupled model (i.e., in the absence of binary interactions between CL molecules, interacting parameter a = 0), which gives a CL intrinsic curvature of c_CL = −1.12 ± 0.4 nm⁻¹. Solid lines represent the minimum square fit to the interacting uncoupled model (i.e., assuming possible CL–CL interactions with the free interacting parameter a is represented with solid lines and gives c_CL = −1.10 ± 0.05 nm⁻¹ and a = (−18 ± 1) k_BT nm². The computations are made with the following values: CL area A_CL = 1.3 nm², bending modulus of a pure CL bilayer, this is κ_CL = 26k_BT and bending modulus of a pure EPC bilayer, κ_EPC = 10k_BT

Fig. 3

CL density in the nanotube’s monolayers according to the uncoupled model. CL molecules with a negative intrinsic curvature (c_CL = −1.1 nm⁻¹) are predicted to be enriched in the inner monolayer (green lines), which bend the membrane in the same sense than the imposed curvature, whereas they are progressively depleted from the outer monolayer (red lines). Unlike the non-interacting model (green dashed line), the interacting model predicts CL enrichment in the inner monolayer of the nanotube (green solid line). The computations are made with the following parametric set (CL area A_CL = 1.3 nm², bending modulus of a pure CL bilayer κ_CL = 26k_BT, bending modulus of a pure EPC bilayer κ_EPC = 10 k_BT, for a CL density in the GUV of ρ_GUV,CL = 0.04 molecules per nm², corresponding to an area fraction of about 6% (half in each monolayer)